Large Vision-Language Models

Also known as LVLM

Follow this topic Ask about this topic

Momentum

30 papers in the last four weeks, up 36% on the four weeks before. 0.3% of all new papers.

Jul 13Week of Sep 28

Latest papers 253

Jun 17, 2026cs.CV

AMALIA-VL: A Native European Portuguese Open-Source Vision and Language Model

Large Vision and Language Models (LVLMs) have advanced rapidly, yet European Portuguese (pt-PT) remains systematically underserved by existing open-source multimodal models, which either conflate it with Brazilian Portuguese or severely under-represent it in their training data mixes. We introduce AMALIA-VL, the first open-source instruction-tuned LVLM built natively for pt-PT, pairing a high-resolution vision encoder with dynamic image tiling and a fully open pt-PT-optimized language model via a learned connector. We contribute with a purposefully designed three-stage training process - vision-language alignment, general visual instruction tuning, and preference optimization - together with a pt-PT-centric multimodal data mix combining curated and translated public datasets with novel datasets that address the near-total absence of European Portuguese multimodal resources. Our evaluation shows that AMALIA-VL establishes a strong baseline for open-source pt-PT LVLMs. We will release model weights, training data, and construction pipelines along with machine-translated pt-PT evaluation benchmarks to help democratize pt-PT LVLM development.

Diogo Glória-Silva, João Cardeira, Manuel Letras da Luz +8

1NOVA School of Science and Technology · 2NOVA LINCS

Vision-Language Models Large Vision-Language Models

Jun 15, 2026cs.CV

Focus When Necessary: Adaptive Routing and Collaborative Grounding for Training-Free Visual Grounding

While Multimodal Large Language Models (MLLMs) excel in cross-modal reasoning, they often struggle to perceive fine-grained details in complex high-resolution images. Recent training-free methods address this through image scaling and localized cropping. However, applying these manipulations indiscriminately introduces computational redundancy for simple queries and can degrade accuracy by truncating essential global context or introducing irrelevant background noise. To this end, we propose LazyMCoT, a dynamic and training-free framework that adaptively allocates visual grounding efforts based on sample difficulty. The framework features an Adaptive Routing mechanism that evaluates predictive uncertainty using first-token statistics from a single forward pass. This efficiently bypasses confident cases while ensuring the recall of difficult samples via conformal calibration. For these challenging cases, a Collaborative Grounding module integrates the inherent cross-modal attention of the model with an external visual expert through a two-stage refinement process. This refinement process generates a precise localized display to recover small or occluded targets. Extensive experiments across diverse benchmarks demonstrate that LazyMCoT rivals training-based approaches by simultaneously improving reasoning accuracy and reducing average inference latency. Our code is availble at https://github.com/TencentBAC/LazyMCoT.

Yifan Wang, Peiming Li, Shiyu Li +5

1Tencent BAC · School of Electronic and Computer Engineering, Peking University · School of Mathematics and Statistics, University of Glasgow +1

Efficient VLM Inference Large Vision-Language Models Vision-Language Grounding Visually Grounded Reasoning

Jun 13, 2026cs.CV

Comparing Human Gaze and Vision-Language Model Attention in Safety-Relevant Environments

Human visual attention plays an important role in how people perceive and respond to environments containing potential risks. This study investigates whether large vision-language models can identify the same regions of a scene that attract human attention in safety-relevant environments. Eye-tracking data were collected from ten participants viewing 33 scene images representing environments with varying levels of potential risk using Pupil Invisible wearable glasses. Gaze coordinates were mapped onto stimulus images to generate population-averaged human gaze heatmaps. In parallel, GPT-4o was prompted through the OpenAI Vision Application Programming Interface (API) to generate spatial predictions of visual attention, which were converted into saliency maps for comparison with human gaze patterns. Spatial alignment between human gaze heatmaps and model-generated saliency maps was evaluated using four complementary metrics: Pearson correlation (r = 0.515 +- 0.117), Normalised Scanpath Saliency (NSS = 0.988 +- 0.323), Kullback-Leibler divergence (KL = 1.766 +- 0.844), and Area Under the Receiver Operating Characteristic Curve using the Judd formulation (AUC-Judd = 0.806 +- 0.076). A cross-model comparison with Gemini Pro, Gemini Flash, and Claude showed that all models exceeded the AUC-Judd chance baseline of 0.5 and achieved positive NSS scores. Gemini Pro demonstrated the strongest spatial localisation according to three of the four metrics, whereas GPT-4o produced the closest distributional match to human attention as measured by KL divergence. These findings suggest that large vision-language models can identify regions that broadly correspond to where humans direct visual attention in safety-relevant scenes without requiring eye-tracking training data. The results highlight the potential of vision-language models as a scalable tool for approximating human attentional patterns.

Marta Vallejo, Siwen Wang

School of Mathematical and Computer Sciences, Heriot-Watt University, Edinburgh, EH14 4AS, United Kingdom

VLM Evaluation Large Vision-Language Models Visual Saliency

Jun 12, 2026cs.CV

Understanding Cross-Modal Contributions in Continual Vision-Language Models: A Theoretical Perspective

Continual vision-language models are commonly addressed through sequential fine-tuning; however, although this paradigm enables adaptation to new environments (tasks), it inherently emphasizes the contribution of previously learned environments (tasks) at the expense of the stability required to preserve previously acquired knowledge. While existing approaches have adequately studied continual learning and catastrophic forgetting in vision-language models (VLMs), the theoretical understanding of modality-specific contributions across a sequence of environments remains largely unexplored. In this paper, we present a new theoretical perspective to understand the cross-modal (vision-language) contributions to consecutive environments. We empirically evaluate our theoretical findings on large VLMs and demonstrate their effectiveness in capturing environment-level cross-modal contributions. Our analysis provides deeper insights into continual VLMs, highlighting their contribution robustness to varying task orders and inter-task similarities, and their improved generalization performance.

Salimeh Sekeh, Mary Wisell

San Diego State University

Continual Learning Large Vision-Language Models Continual Learning for VLMs

Jun 12, 2026cs.CV

One Layer's Trash is Another Layer's Treasure: Adaptive Layer-wise Visual Token Selection in LVLMs

Large Vision-Language Models (LVLMs) have achieved remarkable success across diverse multimodal tasks, yet their practical deployment remains constrained by the computational burden arising from lengthy visual tokens. While visual token pruning has emerged as a promising solution, existing methods suffer from a fundamental limitation: once tokens are pruned at a specific layer, they become inaccessible to all subsequent layers, leading to premature information loss that can compromise model performance. Through empirical studies, we observe that different layers exhibit distinct visual region focus, indicating a varying optimal token subset across layers. Motivated by this insight, we propose Adaptive Layer-wise Visual Token Selection (ALVTS), a novel framework that breaks away from the conventional static token pruning paradigm. ALVTS incorporates a lightweight token selector to identify and route important tokens for further processing, while allowing less important tokens to skip the layer, thus minimizing computational redundancy. These two streams of tokens are seamlessly reintegrated before being fed into subsequent layers, facilitating adaptive compression across the entire model. Grounded in our importance consistency constrained low-rank approximation, the proposed token selection module closely emulates the full attention mechanism, effectively capturing its essential patterns without requiring model retraining. Extensive experiments on LLaVA-1.5, LLaVA-NeXT, and Qwen2.5-VL validate the effectiveness of our method. With an 89% token compression ratio, ALVTS retains 96.7% of the original model's accuracy, achieving a superior efficiency-accuracy trade-off for LVLM inference.

Yongru Chen, Kai Zhang, Zeliang Zong +4

Hikvision Research Institute · East China Normal University · Peking University

Efficient VLM Inference Large Vision-Language Models Visual Token Pruning Visual Token Compression

Jun 10, 2026cs.CV

ECA: Efficient Continual Alignment for Open-Ended Image-to-Text Generation

Incremental Learning (IL) for Open-ended Image-to-Text Generation (OpenITG) enables models to continuously generate accurate, contextually relevant text for new images while preserving previously acquired knowledge. Unlike prior studies, this paper addresses a more practical scenario in which the predominant category of visual data shifts over time as environments evolve. In this context, we introduce a new notion of continual alignment, which incrementally adapts the alignment module within pre-trained VLMs to preserve high-quality cross-modal representations. Based on this idea, we propose Efficient Continual Alignment (ECA), a novel exemplar-free IL approach for OpenITG. The key challenge is enabling the model to acquire new, task-specific features while minimizing interference with the established alignment without accessing raw data from previous tasks. To address this, ECA employs three core mechanisms: a Mixture of Query (MoQ) module that adapts task-specific query tokens, a Fisher Dynamic Expansion (FeDEx) that dynamically expands model structure based on a Fisher Information Matrix (FIM)-based metric, and an embedding dictionary with Dictionary Replay (DR) to retain past knowledge. To evaluate ECA's performance, we construct four new IL OpenITG benchmarks that better reflect real-world scenarios. Experimental results demonstrate that ECA significantly mitigates catastrophic forgetting and improves IL performance compared to baseline methods. Code and benchmarks are available at https://github.com/Snowball0823/ECA.

Jiangtao Kong, Peijun Zhao, Chun-Fu Chen +4

Department of Computer Science, William & Mary, Virginia, USA · Global Technology Applied Research, JPMorgan-Chase, USA · Mohamed bin Zayed University of Artificial Intelligence, Masdar City, Abu Dhabi

Large Vision-Language Models Incremental Learning Vision-Language Alignment Continual Learning for VLMs Catastrophic Forgetting

Jun 8, 2026cs.LG

Density Ridge Selective Prediction for LLM and VLM Hallucination Detection under Calibration Label Scarcity

Hallucination detection in large language and vision-language models is increasingly framed as selective prediction, where a detector assigns a confidence score and abstains when confidence is low. Unsupervised sampling detectors (Semantic Entropy) avoid labels but plateau in quality, while supervised probes attain stronger in-distribution scores yet degrade sharply when calibration labels are scarce. We recover the response manifold of an LLM as the density ridge of a kernel density estimate built on a six-dimensional kinematic feature map of hidden state generation trajectories. A test generation is scored by the negated Euclidean distance from its projected feature point to the nearest ridge vertex, yielding a low-dimensional geometric skeleton of the stochastic output distribution. We evaluate against Semantic Entropy, topological methods, and log-probability on six QA benchmarks (HaluEval-QA, TriviaQA, GSM8K, POPE, ScienceQA, A-OKVQA) using eight text and vision LLMs in a deliberately label-scarce protocol (

n_{\text{cal}}{=}200

queries,

N{=}5

generations). Our ridge-based score beats on AUROC with 5-20 points gain, while demonstrating tempered degradation under calibration-label scarcity.

Nina I. Shamsi

Northeastern University Boston, United States

Hallucination Detection in VLMs Vision-Language Models Audio-Language Model Hallucination Detection Selective Prediction Large Vision-Language Models LLM Hallucination Detection

Jun 5, 2026cs.CV

The Last Visible Pixel: Probing Fine-Scale Perception in Vision-Language Models

Recent vision-language models (VLMs) excel at multimodal understanding and reasoning, yet their fine-grained visual perception remains underexplored. A natural extension of ``How many r are there in Strawberry?'' asks: how small a visual pattern can a VLM reliably perceive? As such, we introduce FineSightBench, a new benchmark that systematically probes this limit by separating perception tasks (pixel-level recognition of letters, shapes, objects) from reasoning tasks (spatial reasoning, counting, ordering over small targets) across controlled scales of 4--48px. Through comprehensive experiments and detailed failure mode analysis on state-of-the-art models, we reveal a sharp dissociation: perception saturates around 12px, while reasoning remains limited even at larger scales, with persistent numeracy and sequence errors. These findings expose fundamental deficiencies in VLMs' fine-scale visual reasoning that demand more rigorous evaluation.

Lujun Li, Lama Sleem, Niccolo Gentile +4

University of Luxembourg · Foyer S.A. · Université Paris-Saclay

VLM Evaluation Large Vision-Language Models VLM Reasoning Fine-Grained Visual Perception

Jun 5, 2026cs.AI

Teaching the Way, Not the Answer: Privileged Tutoring Distillation for Multimodal Policy Optimization

Recent post-training methods, particularly Reinforcement Learning with Verifiable Rewards (RLVR), have significantly enhanced the reasoning ability of Large Vision-Language Models (LVLMs). However, the sparse nature of verifiable rewards provides little token-level supervision for failed rollouts, often leading to inefficient exploration in complex multimodal reasoning tasks. Although policy distillation can offer dense guidance, external teacher based methods introduce substantial computational overhead, while answer conditioned tuning methods may expose answer-level information and induce shortcut-like generation behavior. To address these limitations, we propose PTD-PO, a Privileged Tutoring Distillation Policy Optimization framework for RLVR that provides dense guidance without exposing the answer to the student policy. Specifically, PTD-PO constructs structured privileged hints from spatial attention guidance and intermediate textual reasoning steps, and uses them through in-context learning to produce step-wise token-distribution supervision. The student is still optimized under the original answer-free context, and its failed rollouts are aligned with the hint-augmented reference model at the token-distribution level. To further stabilize distillation under the distribution shift between guided and unguided contexts, we introduce a Top-K Jensen-Shannon divergence objective that focuses alignment on informative token probabilities while reducing memory overhead. Experiments on LVLMs ranging from 2B to 8B parameters show that PTD-PO consistently outperforms RLVR and distillation baselines, mitigates entropy collapse, and improves complex multimodal reasoning performance.

Shizhe Xiang, Ke An, Wenlong Yu +4

1Tianjin University · 2Beijing Institute of Technology · 3Singapore Management University +2

Large Vision-Language Models Multimodal Reasoning Reinforcement Learning with Verifiable Rewards On-Policy Distillation RL for VLMs

Jun 4, 2026cs.AI

Learning Visual Spatial Planning from Symbolic State via Modality-Gap-Aware Self-Distillation

While vision-language models excel at general multimodal understanding, they still struggle with visual spatial planning. We attribute this to a perception-reasoning modality gap: visual planning requires models to infer latent state structures from pixels and then reason over the recovered structure to produce valid actions, whereas symbolic planning directly leverages explicit objects and constraints. This creates dual bottlenecks in visual state recovery and multi-step planning. To address this, we propose MGSD, a two-stage modality-gap-aware self-distillation framework. First, a cold-start grounding stage equips the visual student with reliable state representations, minimizing early perception noise. Second, a privileged teacher transfers planning capabilities via on-policy distillation, using explicit symbolic states to supervise the student's own visual rollout prefixes. Crucially, symbolic data is used strictly during training, leaving inference purely visual. Experiments on visual planning benchmarks show that MGSD consistently improves visual planning across both 4B and 8B backbones, raising the macro average by 19.3% and 18.4%, respectively. The resulting models narrow the gap to symbolic-input upper bounds, while ablations and diagnostics confirm that the improvement comes from both visual state recovery and optimal-path reasoning. These results suggest that modality-gap-aware self-distillation improves not only how models perceive actionable states, but also how they plan over the inferred structure. Code is available at https://github.com/Oranger-l/MGSD.

Haocheng Luo, Jiahui Liu, Ruicheng Zhang +6

1Tsinghua University · 2The Hong Kong University of Science and Technology

Cross-Modal Knowledge Distillation Visual Spatial Reasoning Large Vision-Language Models Self-Distillation

Jun 4, 2026cs.CV

Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction

Video event prediction (VEP) requires models to infer unobserved future states from partial video evidence. Existing video MLLMs usually verbalize intermediate future reasoning in text space: once visual evidence is verbalized, fine-grained motion, geometry, and interaction cues can be lost, leading to plausible but visually ungrounded hallucinations. We introduce Future-L1, an interleaved latent visual reasoning framework that lets an MLLM alternate between language tokens and continuous latent visual spans during autoregressive decoding. To train this capability, we construct Future-L1-50K by selecting examples where future visual hints help prediction and align latent states to future-frame embeddings, then further optimize sampled latent trajectories with LA-DAPO, a latent-aware RL objective with outcome-contrastive and temporal-diversity rewards. Future-L1 achieves new state-of-the-art results on both benchmarks: on FutureBench, it improves Qwen3-VL-8B from 61.0 to 85.4 and exceeds the previous best Video-CoE by 10.4 points; on TwiFF-Bench, it improves the average score from 2.44 to 3.04. These results suggest that future-oriented video reasoning benefits from preserving intermediate visual semantics in latent space rather than translating every reasoning step into text.

Tianxiang Jiang, Linquan Wu, Sheng Xia +5

University of Science and Technology of China · Shanghai AI Laboratory · City University of Hong Kong +4

Large Vision-Language Models Latent Visual Reasoning Video Prediction Visually Grounded Reasoning RL for VLMs

Jun 2, 2026cs.CV

Beyond Encoder Accumulation: Measuring Encoder Roles in Multi-Encoder VLMs

As foundation models scale toward fusing more heterogeneous visual streams, understanding how diverse encoders interact under joint training becomes a prerequisite for principled design. Yet large vision-language models (LVLMs) currently lack the tools to do so, and parameter-efficient encoder configurations remain hard to identify before training. To re-examine encoder roles under joint training, on the 16-benchmark Cambrian-1 suite we retrain and evaluate all 31 non-empty subsets of five common vision encoders under a unified pipeline (~20k GPU-hours total), and report three findings. First, retraining each subset from scratch reveals encoder rankings that differ from those obtained by masking encoders on a fixed checkpoint, including which encoder ranks first overall. Second, we decompose each encoder's contribution into two axes, Capacity, the score an encoder reaches on its own, and Necessity, the drop when it is removed from the full pool. The two axes are not interchangeable. Pairing the two highest-Capacity encoders is suboptimal, while pairing a high-Capacity anchor with an adaptive complement matches the full five-encoder model. Adding further encoders beyond this pair yields only marginal gains. Third, at fixed parameter count, per-encoder pre-projector effective rank explains the residual score variation. The strongest pairs combine an anchor whose rank survives joint training with a complement whose rank expands under it, suggesting that higher-rank, less-collapsed projector inputs correspond to a more favorable optimization regime at the encoder-projector interface. Together, the Capacity-Necessity decomposition and the pre-projector rank analysis, along with comprehensive evaluation through retraining, expose a methodological gap in multi-encoder LVLM design, and offer concrete primitives for closing it.

Wei Ding, Yudong Zhang, Ruobing Xie +3

Tsinghua University Beijing, China · Tsinghua University, Tencent Beijing, China · University of Science and Technology Beijing Beijing, China

VLM Evaluation Vision-Language Models Large Vision-Language Models Multimodal Foundation Models

Jun 2, 2026cs.CV

Steer Where It Matters: Token-Level Visual-Sensitivity Steering for LVLMs Hallucination Mitigation

Large vision language models (LVLMs) have made rapid advancements and are deployed across various applications, yet hallucinations remain a major challenge. Activation steering is appealing due to its minimal training overhead and controllability at inference time. However, we found that during autoregressive decoding, visual conditioning affects token prediction sparsely and locally across decoding steps, and many existing methods that average image-versus-no-image differences over the entire sequence dilute these critical signals, yielding low signal-to-noise ratio steering directions. Additionally, many existing methods apply a fixed steering strength, which misallocates the intervention budget, over-perturbs non-critical tokens, and can cause instability. To address these limitations, we propose Token-Level Visual-Sensitivity Steering (TLVS) for hallucination mitigation. Our approach first extracts token-level steering vectors and refines them, and then applies fine-grained, visual-sensitivity-adaptive steering only where it matters. This lightweight, plug-and-play mechanism requires only minimal training for calibration and can be applied across diverse vision-language models. It modulates the steering strength at each decoding step, selectively suppressing hallucination-prone spans while preserving evidence-grounded content. We evaluate TLVS on several benchmarks, including POPE, AMBER, CHAIR (COCO), MMHal, and HallusionBench, demonstrating consistent improvements over previous steering methods.

Ruipeng Zhang, Zhihao Li, C. L. Philip Chen +1

Guangdong Provincial Key Laboratory of Computational AI Models and Cognitive Intelligence, School of Computer Science & Engineering, South China University of Technology, Guangzhou, China · Engineering Research Center of the Ministry of Education on Health Intelligent Perception and Paralleled Digital-Human, Guangzhou, China · Pazhou Lab, Guangzhou, China.

VLM Hallucination Large Vision-Language Models Activation Steering VLM Hallucination Mitigation

Jun 1, 2026cs.LG

BYORn: Bootstrap Your Own Responses to Defend Large Vision-Language Models Against Backdoor Attacks

Supervised fine-tuning is the predominant approach for adapting autoregressive vision-language models to downstream tasks. Recent work has shown that this paradigm is highly vulnerable to backdoor attacks, and that existing defenses are ineffective in open-ended generation settings. In response, we propose BYORn, a backdoor-robust fine-tuning framework motivated by the observation that poisoned target responses are often semantically implausible given the corresponding image-text inputs and a pretrained model. BYORn identifies such misaligned responses and dynamically replaces them with alternative responses generated by the model, thereby breaking the correlation between triggers and target outputs. The resulting objective gradient corresponds to the gradient of the empirical estimate of the population risk upper bound over the clean data distribution. Empirically, BYORn consistently improves robustness to backdoor attacks while preserving clean-task performance, establishing a new trade-off frontier between generalization and attack success rate. Finally, we demonstrate that BYORn remains effective against adaptive attacks specifically designed to circumvent the proposed defense.

Ivan Sabolić, Marin Oršić, Josip Šarić +1

Faculty of Electrical Engineering and Computing, University of Zagreb, Croatia.

Supervised Fine-Tuning VLM Robustness Large Vision-Language Models Backdoor Attacks on VLMs

Jun 1, 2026cs.CV

ProtoAda: Prototype-Guided Adaptive Adapter Expansion and Geometric Consolidation for Multimodal Continual Instruction Tuning

Multimodal Large Language Models (MLLMs) achieve strong performance through instruction tuning, but real-world deployment requires them to continually acquire new vision-language capabilities, making Multimodal Continual Instruction Tuning (MCIT) essential. To reduce inter-task interference and promote collaboration, recent methods often employ sparse architectures like Mixture of LoRA Experts with image-text similarity routing. However, tasks with distinct response structures could share highly similar visual-linguistic semantics and thus be wrongly routed to the same expert; image-text similarity alone is insufficient for reliable task assignment. For example, an expert in a grounding task requiring coordinate prediction may be biased toward producing short textual answers after learning semantically similar VQA tasks. This format-blind task assignment integrates heterogeneous response types into shared parameters, inducing gradient interference and ineffective expert collaboration. To address this problem, we propose ProtoAda, a prototype-guided adaptive tuning framework. ProtoAda introduces format-aware task prototypes to align task assignment and routing with both task semantics and output structure, and further consolidates format-compatible updates in a geometry-aware manner to effectively reuse and progressively refine existing parameters. Extensive experiments on multiple benchmarks demonstrate that ProtoAda achieves superior performance, especially on tasks whose answer structures are easily corrupted by sequential tuning.

Yu-Cheng Shi, Zhen-Hao Xie, Jun-Tao Tang +1

School of Artificial Intelligence, Nanjing University, China · State Key Laboratory of Novel Software Technology, Nanjing University, China

Large Vision-Language Models Adapter Tuning Continual Learning for VLMs Multimodal Instruction Tuning

Jun 1, 2026cs.CV

EvoCut: Multi-Layer Evolution-Aware Visual Token Compression for Efficient Large Vision-Language Models

Large vision-language models (LVLMs) achieve strong performance on image and video understanding tasks, but their inference efficiency is constrained by the large number of visual tokens produced by vision encoders. Most existing visual token compression methods estimate token importance from attention scores or representation properties at specific layers, overlooking how visual tokens evolve across the vision encoder. Such layer-specific criteria may provide incomplete importance estimates and limit performance preservation after compression. To address this issue, we analyze layer-wise visual token evolution directions and observe that tokens form multiple group evolution directions across vision-encoder layers. Our analysis further shows that informative tokens tend to exhibit persistent deviations from common group evolution directions. Based on this observation, we propose EvoCut, a training-free and attention-free visual token compression method that estimates token importance from multi-layer evolution deviation. Experimental results show that EvoCut can retain only 11.1% of the visual tokens on LLaVA-1.5-7B while preserving 94.4% of the average performance, demonstrating its effectiveness in balancing efficiency and accuracy.

Hongyu Lu, Feng Zhang, Wenwei Jin +5

Harbin Institute of Technology · Xiaohongshu · Fudan University

Efficient VLM Inference Large Vision-Language Models Visual Token Compression

Jun 1, 2026cs.CV

Self-Improving Small Object Grounding in LVLMs

Can internal attention patterns in Large Vision Language Models (LVLMs) identify reliable small-object boxes without fine-tuning? In this work, we provide an affirmative answer. Attention structure in LVLMs encodes grounding quality-a lightweight IoU regressor trained solely on attention maps achieves strong IoU prediction (Pearson r > 0.67). This regressor powers the regressor-based variant of our Attention-based Candidate Selection (ACS) framework, called ACS-Learned, which selects the best box from multiple sampled candidates to improve object grounding. By analyzing what the regressor learns, we reveal which transformer layers and heads are most critical and derive ACS-Free: a training-free selector that ranks candidates by attention entropy on these discriminative heads, with no learned component at inference. Experiments on COCO and Objects365 demonstrate up to 19% self-improvement on small object localization, with ACS-Free ranking best among all training-free methods, demonstrating that useful attention structure improves both localization reliability and interpretability in LVLMs.

Tianze Yang, Yucheng Shi, Ruitong Sun +2

University of Georgia

Large Vision-Language Models VLM Interpretability Vision-Language Grounding Visual Grounding

May 31, 2026cs.CV

An Open-Source Benchmark and Baseline for Multi-temporal Referring Segmentation

Large Vision-Language Models (LVLMs) have shown strong visual understanding and language-guided grounding abilities, yet their capacity for multi-temporal visual reasoning remains underexplored. To bridge this gap, we introduce \textbf{Multi-temporal Referring Segmentation (MTRS)}, a new task that aims to segment language-described temporal changes from multi-temporal images. MTRS extends conventional referring segmentation and change detection by jointly requiring temporal correspondence reasoning, language grounding, and pixel-level mask prediction. We propose \textbf{CRAFT-Agent}, an automated data construction pipeline with human auditing, and build \textbf{MTRefSeg-21K}, the first MTRS benchmark, containing 21K high-quality multi-temporal image-text-mask triplets across diverse scenes, viewpoints, and domains. Benchmarking a broad set of VLM- and LVLM-based models reveals that direct inference performs poorly, while task-specific fine-tuning remains limited. To address this, we propose \textbf{MTRefSeg-R1}, a change-aware LVLM framework trained with a two-stage strategy. It first learns general temporal-change perception from 20K vision-only bi-temporal samples, and is then fine-tuned on MTRefSeg-21K for fine-grained language-guided temporal localization. MTRefSeg-R1 explicitly models cross-temporal visual differences, aligns language instructions with temporal variations, and predicts referred change masks. Extensive experiments show that MTRefSeg-R1 achieves strong and often superior performance compared with existing LVLM baselines, demonstrating the challenge and potential of MTRS.

Bingyu Li, Da Zhang, Tao Huo +3

University of Science and Technology of China, Hefei, China · Institute of Artificial Intelligence (TeleAI), China Telecom, China · School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, China

Large Vision-Language Models Referring Image Segmentation Vision-Language Grounding

May 30, 2026cs.CV

ETC: Extreme Token Compression via Task-aware Visual Information Distillation in VLMs

In Vision-Language Models (VLMs), high-resolution images produce a large number of visual tokens, resulting in high computational costs and KV-cache overhead during inference. To address this problem, we propose an Extreme Token Compression (ETC) framework that minimizes task loss when reducing the number of input tokens based on the principle of variational information distillation. Specifically, from an information-theoretic perspective, we show that minimizing task loss requires the compact representation to preserve the instruction-aware sufficient statistic of the task-relevant visual information for prediction. In practice, ETC leverages text-to-image cross-attention to weight the original visual features to approximate the latent instruction-aware predictive statistic. Moreover, ETC introduces a variational information distillation, enabling the compact representation to preserve the essential information to recover this predictive statistic. Experiments on LLaVA-1.5-7B and Qwen3-VL-2B show that ETC remains effective even under single-token compression, substantially reducing KV-cache overhead while retaining strong task performance.

Yiling Gao, Hongchen Wei, Zhenzhong Chen

School of Remote Sensing and Information Engineering, Wuhan University

Efficient VLM Inference Large Vision-Language Models Multimodal Token Compression Visual Token Compression

May 29, 2026cs.CV

Hyperbolic and Evidence-Prioritized Experts for Large Vision-Language Models

Large Vision-Language Models (LVLMs) have demonstrated impressive performance on multimodal tasks through scaled architectures and extensive training. Recent studies introduce Mixture of Experts (MoE) into LVLMs for improved computational efficiency. However, existing MoE approaches treat visual and linguistic modalities with symmetric architectures, overlooking the inherent asymmetry in how these two modalities are processed. This asymmetry causes two critical issues. First, text and vision form hierarchical rather than parallel relationships, as text queries typically describe partial aspects of complete visual scenes. Euclidean expert space struggles to encode such containment structures. Second, language experts in deeper layers progressively shift from evidence-based processing to parametric memory dependence, losing grounding in the provided visual and linguistic information. To address these issues, we propose AsyMoE, a novel architecture that explicitly models this asymmetry through three specialized expert groups. Intra-modality experts handle modality-specific processing. Hyperbolic inter-modality experts capture hierarchical cross-modal relationships through negative curvature geometry. Evidence-priority language experts suppress parametric memory activation and maintain contextual grounding throughout network depth. Extensive experiments demonstrate that AsyMoE achieves consistent improvements over baseline methods, with average gains of 1.5% over MoE variants and up to 3.8% on hallucination-sensitive tasks. AsyMoE activates 25.45% fewer parameters compared to dense models.

Zijie Zhou, Dandan Zhu, Hangxiangpan Wang +3

China University of Petroleum (Beijing) · Hainan Institute of China University of Petroleum (Beijing) · South China Normal University

Vision-Language Models Large Vision-Language Models VLM Hallucination Mitigation

May 29, 2026cs.CV

SOCO: Benchmarking Semantic Object Correspondence in Vision Foundation Models

Measuring structured object understanding in vision foundation models remains challenging due to inconsistent evaluation protocols and limited part-level supervision. Semantic correspondence (SC) evaluates this capability by testing whether object parts can be matched across instances and categories under large variations in appearance, viewpoint, and geometry. To enable a systematic SC evaluation, we introduce SOCO, a new benchmark for Semantic Object Correspondence that introduces a taxonomy of correspondence types and provides consistent, functionally meaningful keypoint annotations across 100 categories and over 1M correspondence pairs. In addition, SOCO includes keypoint language descriptions, enabling the evaluation of large vision-language models (LVLMs) and their fine-grained part-level understanding. Comprehensive experiments reveal that (i) vision foundation backbones encode strong semantic structure but transfer correspondences poorly across related categories and only partially capture object-part position, (ii) LVLMs are stronger at text-prompted part localization than at visual-reference cross-image matching, exposing a gap between language-grounded localization and fine-grained visual correspondence, and (iii) correspondence performance predicts performance on dense downstream tasks, including segmentation, tracking, 3D pose estimation, and 3D detection, more strongly than ImageNet classification. Together, these findings position SOCO as a benchmark for structured, part-level representation quality in vision and multimodal foundation models.

Olaf Dünkel, Basavaraj Sunagad, Haoran Wang +3

Max Planck Institute for Informatics, SIC · CISPA Helmholtz Center for Information Security · University of Freiburg

VLM Evaluation Semantic Correspondence Vision Foundation Models Large Vision-Language Models Fine-Grained Visual Perception

May 29, 2026cs.CV

Personalize Your Large Vision-language Models With In-context Prompt Tuning

Large vision-language models (LVLMs) have demonstrated strong general multimodal capability and are increasingly deployed in downstream systems. This trend has driven growing interest in LVLM personalization, which aims to enable models to quickly and effectively learn out-of-distribution multimodal concepts to meet user-specific needs. However, many existing methods rely on inference-time training, which reduces efficiency. They also struggle to maintain accuracy in complex multi-image, multi-concept settings. These limitations restrict the broader deployment of LVLM-based systems. Therefore, this paper proposes in-context prompt tuning (ICPT). Specifically, ICPT employs a lightweight projection module capable of operating in complex scenarios to extract fine-grained visual semantics from multiple reference images, seamlessly transforming these features alongside identity-label mappings into continuous prompts. To maximize computational efficiency, this module adaptively determines the prompt length based on the intrinsic visual complexity of each concept. Crucially, to overcome the environmental biases and cross-concept interference prevalent in real-world applications, we introduce two novel geometric regularizations. These constraints refine prompt representations by decoupling key identities from transient environmental states and separating concepts to avoid semantic confusion. Extensive experiments show that ICPT achieves state-of-the-art personalization accuracy across diverse tasks and LVLM backbones.

Yanshu Li, Jiaqian Li, Kuai Yu +4

Brown University · Columbia University · University of Alabama at Birmingham +2

Vision-Language Models Large Vision-Language Models Visual Prompt Learning

May 29, 2026cs.CV

iVGR: Internalizing Visually Grounded Reasoning for MLLMs with Reinforcement Learning

While visually grounded Chain-of-Thought (CoT) has emerged as a promising paradigm to enhance fine-grained perception in multimodal large language models (MLLMs), its efficacy during the inference phase remains underexplored. In this work, we empirically find that mandating explicit object boxes in visually grounded CoT during inference often degrades performance compared to standard textual CoT, which reasons without explicit visual grounding. We hypothesize that the visual localization capability can be internalized into the textual CoT and that the mandatory explicit grounding introduces unnecessary interference with the model's primary objective of answer prediction. To address this problem, we propose Internalizing Visually Grounded Reasoning (\textbf{iVGR}), a novel reinforcement learning framework that transfers localization capabilities into the textual reasoning process. We employ a dual-stream training strategy, where a textual stream is aligned with a high-quality visually grounded stream via a proposed consistency reward, enabling the model to localize accurately without explicit grounding during inference. Extensive experiments demonstrate that our method significantly outperforms existing baselines on fine-grained benchmarks, while maintaining the flexibility to support tool-assisted inference workflows.

Chang-Bin Zhang, Yujie Zhong, Qiang Zhang +1

Visual AI Lab, The University of Hong Kong · Independent Researcher · University of Science and Technology of China

Large Vision-Language Models Vision-Language Grounding Visually Grounded Reasoning Fine-Grained Visual Perception RL for VLMs

May 29, 2026cs.CV

Immuno-VLM: Immunizing Large Vision-Language Models via Generative Semantic Antibodies for Open-World Trustworthiness

Large Vision-Language Models have achieved unprecedented success in zero-shot recognition by aligning visual features with broad semantic concepts. However, this semantic abstraction creates a critical vulnerability in open-world deployment: the

Hubris of Semantics'', where models force-fit unknown anomalies into known categories with high confidence due to the lack of explicit negative knowledge. To address this \textit{Open-World Trustworthiness Paradox}, we propose \textbf{Immuno-VLM}, a bio-inspired framework that adapts the biological principle of \textbf{Immunological Negative Selection} to high-dimensional latent spaces. Departing from traditional Open-Set Recognition methods that rely on passive density estimation or inefficient pixel-space outlier generation, Immuno-VLM leverages the generative reasoning of Large Language Models to actively hallucinate

Semantic Antibodies'', textual descriptions of near-distribution outliers (e.g., look-alikes, contextual anomalies) that effectively bound the decision space of known classes.Extensive experiments on ImageNet-1K and four challenging OOD benchmarks reveal that Immuno-VLM establishes a new state-of-the-art.

Xiang Fang, Wanlong Fang, Wei Ji

School of Software Engineering, Huazhong University of Science and Technology · Nanyang Technological University, Singapore · Nanjing University

VLM Robustness Large Vision-Language Models Open-Set Recognition OOD Detection Zero-Shot Anomaly Detection

May 28, 2026cs.CV

ReGuLaR: Relation-Grounded Latent Reasoning for Large Vision-Language Models

Chain-of-thought (CoT) reasoning has significantly improved the reasoning ability of large vision-language models (LVLMs) by verbalizing intermediate reasoning steps in natural language. However, such discrete textual rationales are often insufficient for encoding continuous visual evidence. Recent work addresses this limitation by moving reasoning into continuous latent space. Despite promising progress, existing methods leave latent reasoning insufficiently connected to the compositional and relational structure of visual evidence. To address this gap, we introduce ReGuLaR, a relation grounded latent reasoning framework that explicitly grounds latent states in these critical yet overlooked visual evidence. ReGuLaR uses a training-time ReGFormer to focus latent reasoning on question-relevant objects and inter-object relations, while at inference time the model reasons and generates answers without invoking the ReGFormer. To support training ReGuLaR, we construct RGROUNDING-351K, a real-world vision-language dataset annotated with key object bounding boxes and inter-object relations. Extensive experiments across diverse benchmarks show that ReGuLaR consistently outperforms existing approaches and achieves state-of-the-art performance. We include our code in the submission and will release the code and training data publicly upon acceptance.

Zihu Wang, Karthik Somayaji N. S, Peng Li

University of California, Santa Barbara

Visual Reasoning Large Vision-Language Models Latent Visual Reasoning Vision-Language Grounding VLM Reasoning Visually Grounded Reasoning

May 28, 2026cs.CV

PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding

Large Vision-Language Models (LVLMs) map visual inputs into dense token sequences, imposing a quadratic computational bottleneck for inference. Elastic visual-token compression addresses this by training a single model that can run at multiple visual-token budgets. However, existing approaches struggle under aggressive compression. Spatial-only compression, as in nested pooling, behaves as an imperfect low-pass filter and induces spectral aliasing that obscures fine-grained detail. Query-only compression, as in nested query resampling, replaces explicit grid-aligned tokens with non-local summaries and substantially degrades spatial grounding. To resolve this representational conflict, we introduce PARCEL (Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding), a visual tokenization architecture that dynamically partitions the labor of feature extraction. PARCEL establishes spatial pool tokens as low-frequency layout anchors and conditions elastic query tokens on these anchors through Pool-Conditioned Query Resampling. This encourages query tokens to focus on complementary visual features rather than redundant spatial mapping. Extensive evaluations across 27 benchmarks show that PARCEL improves the performance-efficiency Pareto frontier, consistently outperforming existing matryoshka baselines across visual-token budgets while preserving the "train once, deploy anywhere" paradigm.

Selim Kuzucu, Alessio Tonioni, Vasile Lup +3

Max Planck Institute for Informatics, SIC · Google · Technical University of Munich

Image Tokenization Efficient VLM Inference Visual Tokenization Large Vision-Language Models Visual Token Compression

May 28, 2026cs.CV

Mitigating Hallucination in Vision-Language Models through Barrier-Regulated Adaptive Closed-form Steering

Large vision-language models (LVLMs) often hallucinate objects that are not present in the input image, largely because visual grounding weakens as decoding progresses. Existing inference-time mitigation methods modify logits or hidden states throughout generation, but they suffer from three key limitations: they lack an explicit grounding objective, intervene even when the model is already well-grounded, and use fixed correction strengths that do not adapt to the severity of grounding failure. We propose BRACS (Barrier-Regulated Adaptive Closed-form Steering), a training-free steering framework that addresses these issues through barrier-regulated adaptive closed-form steering. BRACS monitors the model's own attention to measure visual grounding and applies corrections to the hidden states only when grounding deteriorates. The corrective update is computed analytically in closed form, requiring no training of auxiliary networks or model retraining. Experiments on LLaVA-1.5-7B and Qwen-VL-Chat show that BRACS consistently outperforms prior methods on hallucination benchmarks, reducing CHAIR

_s

by 9.4 points and improving POPE F1 by 2.7 points, while matching or improving performance on four general multimodal benchmarks. BRACS also remains efficient, operating at 80% of greedy decoding throughput and achieving 1.3 times higher speed on average than the baselines.

Soumyadeep Jana, Pulkit Mittal, Sanasam Ranbir Singh

Indian Institute of Technology Guwahati, India

Hallucination in Language Models Large Vision-Language Models Object Hallucination in VLMs Vision-Language Grounding Activation Steering VLM Hallucination Mitigation+1

May 28, 2026cs.CV

OccamToken: Efficient VLM Inference with Training-Free and Budget-Adaptive Token Pruning

Vision-language models (VLMs) rely on long visual token sequences for visual understanding, making the prefill stage expensive in both computation and memory. Most existing pruning methods follow an absolute-ranking paradigm, assigning importance scores to visual tokens and retaining a fixed top-K subset. In this work, we argue that this paradigm is fundamentally brittle: attention sinks distort token importance rankings, while image redundancy and query-dependent visual evidence make fixed token budgets unreliable across inputs. We propose OccamToken, a training-free framework that replaces absolute token ranking with register-anchored relative evidence testing. Instead of asking which tokens are globally important, OccamToken evaluates whether a visual token provides information beyond a register-based reference. Our key insight is that register tokens naturally absorb low-information attention patterns, making them a stable reference for identifying genuinely informative visual evidence. Based on this principle, OccamToken performs both image-adaptive redundancy pruning and query-adaptive relevance pruning through dynamic thresholds derived from register attention. Across LLaVA-NeXT, LLaVA-v1.5, and Qwen3-VL, OccamToken consistently improves the accuracy-efficiency trade-off without additional training. Notably, on LLaVA-NeXT, it reduces 2,880 visual tokens to approximately 40 while preserving over 93% of the original accuracy, enabling stable visual token compression even in the extreme 1.4% retention regime.

Geng Li, Guohao Chen, Ting Chen +6

1Nanyang Technological University (NTU)

Efficient VLM Inference Large Vision-Language Models Visual Token Pruning Visual Token Compression

May 27, 2026cs.CV

Self-Prophetic Decoding to Unlock Visual Search in LVLMs

Large Vision-Language Models (LVLMs) are rapidly evolving toward true multimodal reasoning, with visual search representing a concrete instantiation of the thinking-with-images paradigm. However, LVLM visual search faces two key challenges: incompatibility among intrinsic capabilities after post-training, and interference in long multi-step reasoning contexts. To address these, we identify two novel insights. First, self-regulation between pre- and post-training LVLMs leverages the intrinsic single-step capabilities of the pre-training model to mitigate capability deterioration and long-context interference. Second, probability-based prophetic sampling, replacing naive prompting, provides a probabilistic interface where the pre-training model acts as a prophet and the post-training model selectively accepts prophetic tokens under its output distribution, preserving coherent multi-step reasoning. Building on these insights, we introduce SeProD, a self-prophetic decoding framework that leverages intrinsic single-step capabilities to enable coherent multi-step reasoning in a training-free, plug-and-play manner. Experiments show that SeProD consistently improves multiple visual-search LVLMs across all 12 splits of 4 visual search benchmarks, as well as across general VQA benchmarks, without added computational overhead, thanks to its parallel prophetic acceptance mechanism.

Zhendong He, Qiyuan Dai, Guanbin Li +2

School of Computer Science and Engineering, Sun Yat-sen University · ShanghaiTech University

Large Vision-Language Models VLM Reasoning Visual Search VLM Inference

May 27, 2026cs.CL

Risk-aware Selective Prompting for Hallucination Mitigation in Large Vision-Language Models

Prompt-based verification is widely used to mitigate hallucinations in large vision-language models (LVLMs), yet when it helps remains poorly understood. We systematically study verification prompting across two representative LVLM architectures and hallucination benchmarks, and find that it is a risk-bearing intervention: its corrections increase with input difficulty, while newly introduced errors persist across difficulty levels. As a result, always-on prompting helps on hard inputs but offers little benefit -- and can harm -- easier ones. Our analysis further shows that this behavior is associated with a conservative output shift. Verification prompts redistribute attention from visual tokens toward instruction tokens and induce a distinct middle-layer entropy pattern absent in a neutral-prompt control, suggesting instruction-conditioned attention redistribution rather than uniformly improved visual grounding. Motivated by this input-dependent risk, we propose Risk-aware Selective Prompting (RSP), a training-free approach that uses pre-generation uncertainty signals to trigger verification selectively. RSP mitigates the degradation of always-on prompting while preserving baseline performance, and reveals that effective selection signals vary across architectures.

Yuang Huang, Yafeng Zhang, Yu Zilan

Shanghai Jiao Tong University · iFLYTEK · Tsinghua University

Vision-Language Models Hallucination in Language Models VLM Hallucination Large Vision-Language Models VLM Hallucination Mitigation

May 27, 2026cs.CV

Structure-Guided Visual Perturbation Neutralization for LVLMs

Image inputs enable Large Vision Language Models (LVLMs) to perceive fine-grained visual information, but also introduce a pixel-level attack surface through which adversarial perturbations can elicit unsafe model behaviors. However, most existing defenses are designed for traditional computer vision settings and thus often overlook the cross-modal alignment required by LVLMs, leading to degraded performance. Meanwhile, the limited defenses tailored to LVLMs often require substantial image modifications and introduce considerable computational overhead, thereby compromising inference quality and efficiency. To address these limitations, we propose Structure-Induced Guided Neutralization (SIGN), a lightweight, plug-and-play defense framework that improves LVLM compatibility via Prior Structural Extraction and achieves efficient perturbation suppression via Dynamic Guided Neutralization. Extensive experiments show that SIGN achieves over 87% defense success rate with only 0.5% pixel modification and 0.16 seconds per image, while nearly preserving original visual representations and benign task performance. Our work offers a lightweight alternative to defenses that require costly model training and highlights the potential of exploiting a vision encoder for efficient adversarial protection. Our code is open source on https://anonymous.4open.science/r/SIGN-BCB1.

Yuanhe Zhang, Xueting Wang, YanBin Ren +6

1Beijing University of Posts and Telecommunications · University of Science and Technology of China · 3JIUTIAN Research +2

VLM Robustness Adversarial Attacks on VLMs Large Vision-Language Models Adversarial Examples

May 26, 2026cs.CV

Leveraging Visual Signals for Robust Token-Level Uncertainty in Vision-Language Generation

Uncertainty quantification (UQ) remains a critical challenge in Large Vision Language Models (LVLMs) for reliable predictions and real-world deployment. However, most existing methods are adapted from the LLM literature and primarily focus on the language modality, leaving the contribution of visual information to LVLM uncertainty largely underexplored. In this paper, we investigate how LVLMs process visual information and whether this process can be used to improve uncertainty estimation. By analyzing hidden representations after the integration of visual features during the generation process, we observe that high-confidence predictions rely more heavily on visual content than uncertain ones. Building on this insight, we propose Visual-Grounded Token UQ (VIG-TUQ), a training-free framework that explicitly incorporates visual grounding into uncertainty estimation by weighting token-level language uncertainty with visual grounding scores. We evaluate VIG-TUQ on multiple datasets and across diverse LVLM architectures, including early-fusion, late-fusion, and native-fusion models. Results indicate that our method often improves upon existing token-level uncertainty approaches. Code and data will be made available upon acceptance.

Joseph Hoche, David Brellmann, Gianni Franchi

AMIAD, Pôle Recherche, Palaiseau · Safran Tech · ENSTA Paris

Vision-Language Models Large Vision-Language Models LLM Uncertainty Estimation

May 25, 2026cs.CV

LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence

We introduce LLaVA-OneVision-2 (LLaVA-OV-2), the most capable vision-language model in the LLaVA-OneVision series to date, achieving superior performance across a broad range of multimodal benchmarks. The model builds on a native OneVision-Encoder and incorporates Windowed Attention for efficient local computation while maintaining native resolution. Its key advance is codec-stream tokenization: it treats compressed video as a continuous bit-cost stream, where bit-cost dynamics determine adaptive temporal groups, and motion-residual cues select salient spatial evidence into compact visual canvases. This allocation concentrates a limited token budget on event-bearing content, enabling more stable long-video token compression than fixed groups of pictures. A shared 3D RoPE further places codec canvases, sampled frames, and images in a unified spatiotemporal coordinate system. Furthermore, we build the LLaVA-OV-2 data and training stack around large-scale open supervision: approximately 8M re-captioned video samples for pretraining, a 4M-sample spatial corpus for fine-tuning. We also introduce JumpScore, a temporal-localization benchmark targeting fine-grained grounding in high-frequency, densely repeated motion, a regime underrepresented by existing video evaluations. A standout capability of LLaVA-OV-2 is its unified perception across video understanding, temporal grounding, spatial grounding, and manipulation-trace reasoning. On JumpScore, LLaVA-OneVision-2-8B reaches 74.9 JumpScore mAP, surpassing Qwen3-VL-8B (30.1) by +44.8 points; under matched visual-token budgets on the same benchmark, codec-stream inputs improve temporal grounding over frame sampling by +9.7 points. Across standard benchmarks, LLaVA-OneVision-2-8B further outperforms Qwen3-VL-8B by +4.3 average points on video tasks, +5.3 on spatial tasks, and +15.6 average J&F on tracking tasks.

Xiang An, Yin Xie, Feilong Tang +27

Glint Lab · AIM for Health Lab · MVP Lab

VLM Evaluation Temporal Video Grounding Visual Tokenization Large Vision-Language Models Video Token Compression

May 25, 2026cs.CV

Beyond Appearance: Can Multimodal Large Language Models Exploit Vertical Structure for Remote Sensing Natural Scene Understanding?

Multimodal large language models (MLLMs) have advanced rapidly in remote-sensing analysis, yet existing evaluations remain predominantly 2D-centric. Because spectrally confused regions can appear nearly identical yet differ substantially in vertical structure, appearance alone is often insufficient for reliable semantic interpretation in natural scenes. Vertical structure therefore provides decision-critical physical evidence, yet whether current MLLMs can effectively perceive, ground, and utilize such geometric evidence remains underexplored. To bridge this gap, we introduce VertiCue-Bench, the first diagnostic benchmark that uses controlled interventions to probe whether vertical height evidence is actually perceived, grounded, and utilized, and we establish a three-stage evidence-utilization framework of Perception--Grounding--Utilization. By constructing a Representation Intervention Spectrum spanning multiple presentation and interaction modalities, including Raw Visual, Tool-assisted, and Oracle Text conditions, together with controlled counterfactual tests, we conduct an in-depth disentangled diagnosis across 10 state-of-the-art models. Our experiments reveal and formally characterize the Vertical Structure Utilization Gap. Although current models exhibit emerging geometric perception capabilities, they still struggle to accurately ground vertical evidence to relevant spatial entities and integrate it into high-level semantic decisions. This finding identifies a critical bottleneck in developing physically grounded and geometry-aware remote-sensing MLLMs.

Jing Huang, Duanchu Wang, Junjie Yang +5

School of Software Engineering, Xi’an Jiaotong University, Xi’an 710049, China · Shenyang Institute of Automation, University of Chinese Academy of Sciences, Shenyang, China · State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi’an Jiaotong University

VLM Evaluation Remote Sensing Image Understanding Large Vision-Language Models VLM Reasoning Remote Sensing Benchmarks

May 25, 2026cs.CV

Adversarial Orthogonal Disentanglement for LVLM Hallucination Mitigation

Large Vision-Language Models (LVLMs) have advanced multimodal understanding, yet their reliability is limited by hallucination, where generated content conflicts with visual facts. Existing mitigation methods either rely on costly external interventions, such as instruction tuning and retrieval, or use internal mechanisms that remain limited by flawed attention weights and entangled hidden representations. We propose Adversarial Orthogonal Disentanglement (AOD), a latent geometric framework for mitigating LVLM hallucinations. AOD learns a hallucination-related direction through a minimax objective: a classifier concentrates hallucination signals into the projected component, while an adversary removes them from the orthogonal residual space via a Gradient Reversal Layer. The learned direction enables a training-free dual-forward-pass contrastive decoding strategy that suppresses hallucinations while preserving general capabilities. Experiments on three LVLMs across four hallucination and four utility benchmarks show that AOD consistently outperforms strong baselines. It improves POPE accuracy by over 6% on average, boosts AMBER by 6%, and maintains strong performance on utility tasks such as MMMU. Further analysis shows robust transfer across datasets, suggesting that AOD captures general hallucination-related biases rather than dataset-specific artifacts. Our source code and datasets are available at https://github.com/Hunter-Wrynn/AOD.

Ruoxi Cheng, Haoxuan Ma, Zhengfei Hai +6

Fudan University · Tencent · Nanjing University +3

Adversarial Representation Learning Vision-Language Models Hallucination in Language Models VLM Hallucination Large Vision-Language Models Representation Disentanglement+1

May 23, 2026cs.CV

VaaWIT: Visual-Aware Adaptation of Large Language Models for Multilingual Web Image Translation

Translating text embedded in Web images is crucial for improving content accessibility and cross-lingual information retrieval, particularly within social media and e-commerce domains. Although Large Vision-Language Models (LVLMs) have advanced multimodal understanding, applying them to Web image translation remains challenging due to the visual representation gap: standard encoders often prioritize high-level semantics over the fine-grained visual details required for recognizing diverse character morphologies. To address this challenge, we propose VaaWIT, an end-to-end framework that adapts Large Language Models for multilingual Web image translation. The framework introduces two key technical contributions: (1) a Dual-Stream Attention Module (DSAM), which facilitates bidirectional interaction between multilingual semantic features and detailed visual representations, thereby synthesizing unified features robust to textual variations; and (2) a Visual-Aware Adapter (VAA), a parameter-efficient fine-tuning strategy that dynamically injects these fused visual cues into the frozen LLM backbone. This design enables the model to align the visual context with linguistic reasoning effectively while minimizing computational costs. Extensive experiments on eight tasks on three public benchmarks demonstrate that VaaWIT significantly outperforms state-of-the-art (SOTA) open-source baselines and achieves competitive performance against proprietary models. These results validate the efficacy of integrating fine-grained visual perception into LLMs for complex Web content analysis.

Bo Li, Ronghao Chen, Ningyuan Deng +3

Tsinghua University Baidu Inc. Beijing, China · QuantaAlpha Beijing, China · The Hong Kong University of Science and the Technology Hong Kong, China +1

VLM Adaptation Large Vision-Language Models Machine Translation

May 22, 2026cs.CV

PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs

Despite remarkable progress in Multimodal Large Language Models (MLLMs), these models still struggle with fine-grained understanding tasks. In this work, we propose Procedurally Generated Tasks (PGT), a simple data-driven framework that serves a dual purpose: inducing fine-grained visual understanding and acting as a low-cost diagnostic tool to identify the source of perception failures. By overlaying unambiguous geometric primitives on images, PGT generate additional dense supervision that disentangles visual grounding capability from semantic priors. Extensive experiments on relational, quantitative, and 3D/depth understanding benchmarks show that PGT yields remarkable gains across diverse architectures. Instruction tuning MLLMs on LLaVA-v1.5-Instruct augmented with PGT data results in improvements of up to +20% on the What'sUp benchmark and +13.3% on CV-Bench-2D, while maintaining general perception capabilities. Moreover, finetuning state-of-the-art MLLMs on PGT data leads to boosts of up to +5.5% on What'sUp and +8.3% on CV-Bench-2D. These findings demonstrate that PGT effectively address the bottleneck of fine-grained perception, revealing that many spatial reasoning deficits stem from inadequate supervision signals rather than inherent architectural or resolution limitations.

Rim Assouel, Amir Bar, Michal Drozdzal +1

1Mila - Québec AI Institute · 4McGill University · 3FAIR at Meta Superintelligence Labs +2

Visual Spatial Reasoning Synthetic Data Generation Large Vision-Language Models Vision-Language Grounding Visually Grounded Reasoning Fine-Grained Visual Perception

May 22, 2026cs.CV

CVSearch: Empowering Multimodal LLMs with Cognitive Visual Search for High-Resolution Image Perception

High-resolution (HR) image perception presents a key bottleneck for multimodal large language models (MLLMs). While visual search offers a promising solution, existing methods struggle with the trade-off between coverage and efficiency. Visual expert-assisted search is efficient but prone to blind spots when proposals fail, whereas scan-based search guarantees coverage at the cost of computational redundancy and semantic fragmentation. To address this dilemma, we introduce CVSearch, a training-free adaptive framework that dynamically schedules search strategies via an Assess-then-Search workflow. Specifically, CVSearch first invokes expert-assisted search when global information is insufficient, and only triggers a novel semantic-aware scanning mechanism upon failure. Distinct from rigid grid partitioning, this efficient scanning paradigm incorporates Semantic Guided Adaptive Patching to decompose images into semantically consistent regions, effectively mitigating object fragmentation. Furthermore, we devise a Dynamic Bottom-Up Search strategy driven by a Visual Complexity prior to enable efficient and precise iterative exploration of local details. Extensive experiments on HR benchmarks demonstrate that CVSearch achieves state-of-the-art accuracy while substantially improving search efficiency. Code is released at https://github.com/liliupeng28/ICML26-CVSearch.

Liupeng Li, Haoqian Kang, Zhenyu Lu +4

Harbin Institute of Technology, Shenzhen, China · Peng Cheng Laboratory, Shenzhen, China · Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China.

Efficient VLM Inference Large Vision-Language Models Visual Search

May 21, 2026cs.CV

Exposing Vulnerabilities in Visible-Infrared VLMs: A Unified Geometric Adversarial Framework with Cross-Task Transferability

Vision-language models (VLMs) have achieved strong performance across diverse multimodal tasks, but their adversarial robustness in visible-infrared (VIS-IR) scenarios remains underexplored. This gap is critical because VIS-IR sensing is widely used in real-world perception systems to support reliable understanding under challenging imaging conditions. To address this cross-modal threat setting, we propose CFGPatch, a curved-edge fractal geometric adversarial patch framework for attacking VIS-IR VLMs. CFGPatch builds on triangular fractal geometry and replaces rigid straight-edged primitives with Bezier-curved elements, preserving multi-scale fractal self-similarity while introducing smoother contours, richer directional variation, and more flexible shape deformation. In addition, we design a modality-specific Fraser-spiral rendering mechanism to inject fine-grained texture distortions and misleading perceptual cues into visible and infrared images. By coupling global curved-fractal geometry with local spiral-based appearance interference, CFGPatch disrupts both shape perception and texture interpretation. We further adopt expectation over transformation (EOT) to improve robustness against common image-level transformations. Extensive experiments show that CFGPatch effectively fools VIS-IR VLMs and consistently outperforms standard patch baselines in attack effectiveness and robustness. Moreover, adversarial samples optimized for zero-shot classification transfer well to image captioning and visual question answering, demonstrating strong cross-task transferability and generalizability across downstream tasks.

Xiang Chen, Yuxian Dong, Chao Li +6

VLM Robustness Adversarial Patch Attacks Adversarial Attacks on VLMs Large Vision-Language Models Transfer-Based Adversarial Attacks Physical Adversarial Attacks

May 20, 2026cs.CV

Reducing Object Hallucination in LVLMs via Emphasizing Image-negative Tokens

Object hallucination is a significant challenge that hinders the application of large vision-language models (LVLMs) in practice. We hypothesize that one possible origin of hallucination is the model's tendency to prioritize text generation over meaningful interaction with images. To explore this, we examine the generation process and categorize text tokens into three groups: image-positive, invariant, and negative, based on their visual dependence on input image tokens. Our analysis reveals that most generated tokens are minimally influenced by the image information. This suggests that during the model's training stage, more emphasis is placed on learning how to follow textual instructions, rather than extracting information from images. Based on this finding, we propose adjusting the training weights of different tokens depending on their visual dependence to control hallucination. Additionally, we remove a portion of the training data that potentially contains more hallucinations as a data filtering strategy. Both methods achieve a reduction in hallucination without compromising response length or introducing additional computational costs during inference. We validate our methods across three LVLM variants, demonstrating the effectiveness and general applicability.

Meng Shen, Minghao Wu, Deepu Rajan

Nanyang Technological University · Monash University

Vision-Language Models Hallucination in Language Models Large Vision-Language Models Object Hallucination in VLMs Adaptive Loss Weighting VLM Hallucination Mitigation

May 20, 2026cs.CV

Mitigating Hallucinations in Large Vision-Language Models via Causal Route Gating

Large vision-language models (LVLMs) often hallucinate content that is fluent yet unsupported by the image, limiting their reliability in real-world deployment. We show that a key failure mode arises from route competition: even when visual tokens receive attention, the final token decision can be dominated by the textual pathway, causing the decoder to follow linguistic priors over visual evidence. To mitigate this, we propose a training-free, decision-aligned intervention that decomposes each attention head into a visual route and a text route, and estimates their token-level effects using an efficient one-forward/one-gradient approximation. These estimates reveal route conflict within heads and identify prior-dominant ones, enabling selective suppression of only the text route while keeping the visual route intact. Across five benchmarks spanning discriminative and generative settings, our method consistently reduces hallucination-related errors across models with limited impact on overall multimodal performance, while incurring a modest inference-time overhead.

Zhe Cheng, Wenyu Chen, Fode Zhang +1

Center of Statistical Research, School of Statistics and Data Science, Southwestern University of Finance and Economics, Chengdu, China. · Department of Biomedical Engineering, College of Design and Engineering, National University of Singapore, Singapore.

VLM Robustness Hallucination in Language Models VLM Hallucination Large Vision-Language Models Faithfulness in VLMs VLM Hallucination Mitigation

May 20, 2026cs.CV

Finding the Correct Visual Evidence Without Forgetting: Mitigating Hallucination in LVLMs via Inter-Layer Visual Attention Discrepancy

Large Vision-Language Models (LVLMs) have shown remarkable performance on a wide range of vision-language tasks. Despite this progress, they are still prone to hallucination, generating responses that are inconsistent with visual content. In this work, we find that LVLMs tend to hallucinate when they pay insufficient attention to the correct visual evidence and gradually forget it during the generation process. We empirically find that although LVLMs overall attend insufficiently to visual evidence, they exhibit sensitivity to the correct visual evidence in specific layers, with notable inter-layer discrepancy. Motivated by this observation, we propose a novel hallucination mitigation method that enhances visual evidence based on Inter-Layer Visual Attention Discrepancy (ILVAD). Specifically, we obtain the attention weights from early generated tokens to visual tokens across layers and identify the tokens that are repeatedly activated as visual evidence, forming a saliency map. We then enhance attention to visual evidence during generation through the saliency map to reduce visual forgetting. In addition, we leverage the saliency map to obtain attention scores of generated text to visual evidence, in order to select and emphasize text tokens that are strongly grounded in visual evidence. Our method is training-free and plug-and-play. Multiple benchmark evaluations conducted on five recently released models show that our method can consistently mitigate hallucinations in different LVLMs over various architectures. Code is available at https://github.com/ytx-ML/ILVAD.

Yutong Xie, Zhenglin Hua, Ran Wang +3

School of Computer Science and Engineering, Southeast University, Nanjing, China · School of Artificial Intelligence, Shenzhen University, Shenzhen, China · National Engineering Laboratory for Big Data Systems Computing Technology, Shenzhen University, Shenzhen, China +3

Vision-Language Models Hallucination in Language Models VLM Hallucination Large Vision-Language Models VLM Hallucination Mitigation

May 19, 2026cs.CV

Breaking Modality Heterogeneity in Low-Bit Quantization for Large Vision-Language Models

Low-bit post-training quantization (PTQ) is a pivotal technique for deploying Vision-Language Models (VLMs) on resource-constrained devices. However, existing PTQ methods often degrade VLMs' accuracy due to the heterogeneous activation distributions of text and vision modalities during quantization. We find that this cross-modal heterogeneity is distributed unevenly across channels: a small subset of channels contains most modality-specific outliers, and these outliers typically reside in different channels for each modality. Motivated by this, we propose SplitQ, a channel-Splitting-driven post-training Quantization framework. At its core, SplitQ introduces a novel Modality-specific Outlier Channel Decoupling (MOCD) module that effectively isolates salient modality-specific outlier channels with minimal overhead. To further address the remaining cross-modal distribution discrepancies, we design an Adaptive Cross-Modal Calibration (ACC) module that employs dual lightweight learnable branches to dynamically mitigate modality-induced quantization errors. Extensive experiments on popular VLMs demonstrate that SplitQ significantly outperforms existing approaches across 6 popular multi-modal datasets under all evaluated quantization settings, including W4A8, W4A4, W3A3, and W3A2. Notably, SplitQ preserves 93.5% of FP16 performance under the challenging W3A3 setting (69.5 vs. 74.3), pushing the efficiency frontier for deploying advanced VLMs. Our code is available at https://github.com/EMVision-NK/SplitQ

Yi Zhong, Haotong Qin, Xindong Zhang +2

VCIP, College of Computer Science, Nankai University · D-ITET, ETH Zürich · OPPO Research Institute +1

Vision-Language Models VLM Quantization Large Vision-Language Models Post-Training Quantization Low-Bit Quantization

May 18, 2026cs.CV

Why We Look Where We Look: Emergent Human-like Fixations of a Foveated Visual Language Model Maximizing Scene Understanding

When humans view scenes without a specific task (free-viewing), they initially direct their eye movements toward the scene center and then fixate on people, text, objects being gazed at or grasped, and semantically meaningful regions. What these signature fixation patterns reflect and whether they optimize an underlying perceptual task remain unknown. We show that a computational agent with simulated foveation, trained to optimize scene comprehension, exhibits emergent human fixation signature patterns. In contrast, versions of the agent trained to search or classify scenes, or equipped with peripheral vision that was better or worse than human vision, predicted human fixation patterns less accurately. Thus, human free-viewing fixation patterns may emerge as a functional byproduct of optimizing scene comprehension under the biological constraints of foveated vision.

Shravan Murlidaran, Ziqi Wen, Sana Shehabi +1

Psychological & Brain Sciences, University of California, Santa Barbara & 93106, USA. · Computer Science, University of California, Santa Barbara & 93106, USA. · Electrical and Computer Engineering, University of California, Santa Barbara & 93106, USA.

Visual Attention Large Vision-Language Models Active Visual Perception

May 17, 2026cs.LG

When a Zero-Shooter Cheats: Improving Age Estimation via Activation Steering

Different age-related regulations have been proposed to protect minors from harmful content and interactions online. Automated age estimation is central to enforcing such regulations, and vision-language models (VLMs) achieve state-of-the-art performance on this task. However, we find that the zero-shot nature of VLM-based age estimation produces an unexpected side effect we call the identity shortcut: Instead of estimating age from visual features, VLMs tend to identify the depicted person and infer their age from memorized knowledge. This phenomenon leads to substantially incorrect predictions when non-celebrities are misidentified as celebrities. It also produces deceptively high robustness to noise and adversarial perturbations on celebrity images, which dominate popular benchmarks. To mitigate this, we propose an activation steering method that suppresses the shortcut by intervening on the hidden states of the VLM. This method improves age estimation accuracy for both memorized and unseen identities, reducing mean absolute error by up to 25% across popular benchmarks.

Erik Imgrund, Pia Hanfeld, Klim Kireev +1

BIFOLD & TU Berlin

VLM Robustness Large Vision-Language Models Shortcut Learning

May 16, 2026cs.CV

DriveSafe: A Framework for Risk Detection and Safety Suggestions in Driving Scenarios

Comprehensive situational awareness is essential for autonomous vehicles operating in safety-critical environments, as it enables the identification and mitigation of potential risks. Although recent Multimodal Large Language Models (MLLMs) have shown promise on general vision-language tasks, our findings indicate that zero-shot MLLMs still underperform compared to domain-specific methods in fine-grained, spatially grounded risk assessment. To address this gap, we propose DriveSafe, a framework for risk-aware scene understanding that leverages structured natural language descriptions. Specifically, our method first generates spatially grounded captions enriched with multimodal context, including motion, spatial, and depth cues. These captions are then used for downstream risk assessment, explicitly identifying hazardous objects, their locations, and the unsafe behaviors they imply, followed by actionable safety suggestions. To further improve performance, we employ caption-risk pairings to fine-tune a lightweight adapter module, efficiently injecting domain-specific knowledge into the base LLM. By conditioning risk assessment on explicit language-based scene representations, DriveSafe achieves significant gains over both zero-shot MLLMs and prior domain-specific baselines. Exhaustive experiments on the DRAMA benchmark demonstrate state-of-the-art performance, while ablation studies validate the effectiveness of our key design choices. Project page: https://cvit.iiit.ac.in/ research/projects/cvit-projects/drivesafe

Sainithin Artham, Shankar Gangisetty, Avijit Dasgupta +1

IIIT-Hyderabad, India

VLMs for Autonomous Driving Autonomous Driving Safety Evaluation Large Vision-Language Models Autonomous Driving Autonomous Driving Perception

May 15, 2026cs.CR

A Cross-Modal Prompt Injection Attack against Large Vision-Language Models with Image-Only Perturbation

Large vision-language models (LVLMs) have emerged as a powerful paradigm for multimodal intelligence, but their growing deployment also expands the attack surface of prompt injection. Despite this growing concern, existing attacks still suffer from a critical limitation: the injected prompt for one modality only steers the model's interpretation of that singular input. Alternatively, these attacks remain multimodal but fail to achieve cross-modal prompt perturbation. To bridge this gap, we introduce a novel cross-modal prompt injection attack CrossMPI, which can steer the model's interpretation of both textual and visual inputs via image-only prompt injection. Our design is underpinned by the following key breakthroughs. First, we turn the focus of the injected prompt perturbation optimization from the visual embedding space (typically with only

10^5

parameters) to the model hidden state space (for multimodal information integration and with

10^7

parameters). Then, two strategies are adopted to mitigate the optimization challenges posed by the larger parameter space. To constrain the optimized model parameter space, we introduce a layer selection strategy that identifies the layers most critical to multimodal integration. Interestingly, deviating from the past experience, our analysis reveals that the optimal layers for LVLM prompt perturbation reside in the middle of the model rather than the last. To constrain the image perturbation space, we propose a new distance-decremental perturbation budget assignment strategy that allocates budgets decrementally as the pixel distance to semantic-critical regions increases. Extensive experiments across multiple LVLMs and datasets show that our method significantly outperforms baseline approaches.

Hao Yang, Zhuo Ma, Yang Liu +3

Xidian University Xi’an, Shaanxi, China

VLM Robustness Adversarial Attacks on VLMs Large Vision-Language Models

May 15, 2026cs.CV

EntropyScan: Towards Model-level Backdoor Detection in LVLMs via Visual Attention Entropy

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across various tasks, yet they remain vulnerable to backdoor attacks. Existing defense methods predominantly focus on sample-level defense, which relies on the knowledge of training data or triggers. However, identifying whether a given model is backdoored remains a critical but unexplored task. To fill this gap, we propose EntropyScan, a lightweight and trigger-agnostic method for model-level backdoor detection in LVLMs. We first observe that backdoor injection disrupts the cross-modal alignment, resulting in pronounced structural anomalies in visual attention allocation on benign samples. Based on this insight, EntropyScan detects the backdoor models by quantifying such attention deviations. Specifically, it extracts visual attention distributions from the initial layers of the Large Language Model (LLM) and applies Tsallis entropy to capture these structural distortions. By employing a reference-anchored Z-score normalization on a small set of benign samples, it effectively identifies the backdoored model. Extensive experiments across two LVLMs architectures and three advanced attack scenarios show that EntropyScan achieves an F1 score of 98.5% in average and an AUC of 96.6%. Our code will be publicly available soon.

Xuanyu Ge, Zhongqi Wang, Jie Zhang +2

China University of Geosciences · University of the Chinese Academy of Sciences · Institute of Computing Technology, Chinese Academy of Sciences

Large Vision-Language Models Backdoor Detection Backdoor Attacks on VLMs

May 15, 2026cs.CV

LRCP: Low-Rank Compressibility Guided Visual Token Pruning for Efficient LVLMs

Large vision-language models (LVLMs) achieve strong multimodal understanding, but their inference cost grows rapidly with the number of visual tokens, especially for high-resolution images and long videos. Existing attention-based methods estimate token importance from attention scores, which may introduce positional bias, while representation-based methods reduce visual redundancy based on feature relations or reconstruction errors, overlooking the global structure of the visual token set. In this paper, we revisit visual token compression from the perspective of low-rank compressibility. Across models and datasets, we observe that visual token representations exhibit a pronounced low-rank structure, with a dominant subspace that remains stable even after a large fraction of tokens is randomly removed. Motivated by this finding, we propose LRCP, a training-free compression framework that first estimates the dominant low-rank subspace of visual tokens via PCA, and then scores each token by its projection residual onto this subspace, retaining tokens that are poorly explained by the low-rank background. Extensive experiments show that LRCP achieves superior results, preserving 94.7% of the original image-understanding performance with an 88.9% token reduction and 97.8% of the average video-understanding accuracy with an 87.5% token reduction.

Hongyu Lu, Feng Zhang, Wenwei Jin +5

1Xiaohongshu · 2Harbin Institute of Technology · 3Fudan University

Efficient VLM Inference Large Vision-Language Models Low-Rank Compression Visual Token Pruning Visual Token Compression

May 15, 2026cs.CV

AGC: Adaptive Geodesic Correction for Adversarial Robustness on Vision-Language Models

Vision-language models like CLIP have demonstrated remarkable zero-shot transfer capabilities. However, their susceptibility to imperceptible adversarial perturbations remains a critical security concern. While test-time defenses offer a pragmatic solution for deployed models, existing approaches typically rely on gradient-based optimization during inference, incurring significant computational overhead. In this paper, we revisit the role of data augmentation in CLIP robustness and observe that augmentations are not equally effective: specific augmentations consistently provide robust geometric cues that align with correct class semantics in the hyperspherical feature space. Based on this, we propose Adaptive Geodesic Correction (AGC), a training-free defense mechanism that requires no parameter updates. AGC identifies a reliable augmentation as a geometric anchor and corrects the input feature towards it, utilizing an adaptive step size to balance robustness against clean accuracy preservation. AGC achieves superior performance across eight fine-grained datasets and three CLIP backbones, improving average robust accuracy by 44.4% over state-of-the-art baseline while delivering a 10

\times

reduction in inference latency. Our findings reveal a fundamental geometric property of CLIP features, offering a highly efficient and effective paradigm for robust multimodal deployment.

Zhiwei Li, Jiacheng Xue, Weining Wang +4

1NLPR & MAIS, Institute of Automation, Chinese Academy of Sciences · School of Computer Science and Engineering, Central South University · University of Chinese Academy of Sciences

VLM Robustness Adversarial Robustness Large Vision-Language Models

May 14, 2026cs.CV

Deep Pre-Alignment for VLMs

Most Vision Language Models (VLMs) directly map outputs from ViT encoders to the LLM via a lightweight projector. While effective, recent analysis suggests this architecture suffers from an alignment challenge: visual features remain distant from the text space in the initial layers of the LLM, forcing the model to waste critical depth~\cite{zhang-etal-2024-investigating,artzy-schwartz-2024-attend} on superficial modality alignment rather than deep understanding and complex reasoning. In this work, we propose Deep Pre-Alignment (DPA), a novel architecture that replaces the standard ViT encoder with a small VLM as perceiver, ensuring visual features are deeply aligned with the text space of the target large language model. Comprehensive experiments demonstrate the effectiveness of DPA. On the 4B parameter scale, DPA outperforms baselines by 1.9 points across 8 multimodal benchmarks, with gains widening to 3.0 points at the 32B scale. Moreover, by offloading alignment to the perceiver, DPA achieves a 32.9% reduction in language capability forgetting over 3 text benchmarks. We further demonstrate that these gains are consistent across different LLM families including Qwen3 and LLaMA 3.2, highlighting the generality of our approach. Beyond performance, DPA also offers a seamless upgrade path for current VLM development, requiring only a modular replacement for the visual encoder with marginal computation overhead.

Tianyu Yu, Kechen Fang, Zihao Wan +5

1Tsinghua University · 2Taobao & Tmall Group of Alibaba · 3Shanghai Qi Zhi Institute.

Vision-Language Models Large Vision-Language Models Vision-Language Alignment

May 14, 2026cs.CV

ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both

Visual reasoning, often interleaved with intermediate visual states, has emerged as a promising direction in the field. A straightforward approach is to directly generate images via unified models during reasoning, but this is computationally expensive and architecturally non-trivial. Recent alternatives include agentic reasoning through code or tool calls, and latent reasoning with learnable hidden embeddings. However, agentic methods incur context-switching latency from external execution, while latent methods lack task generalization and are difficult to train with autoregressive parallelization. To combine their strengths while mitigating their limitations, we propose ATLAS, a framework in which a single discrete 'word', termed as a functional token, serves both as an agentic operation and a latent visual reasoning unit. Each functional token is associated with an internalized visual operation, yet requires no visual supervision and remains a standard token in the tokenizer vocabulary, which can be generated via next-token prediction. This design avoids verbose intermediate visual content generation, while preserving compatibility with the vanilla scalable SFT and RL training, without architectural or methodological modifications. To further address the sparsity of functional tokens during RL, we introduce Latent-Anchored GRPO (LA-GRPO), which stabilizes the training by anchoring functional tokens with a statically weighted auxiliary objective, providing stronger gradient updates. Extensive experiments and analyses demonstrate that ATLAS achieves superior performance on challenging benchmarks while maintaining clear interpretability. We hope ATLAS offers a new paradigm inspiring future visual reasoning research.

Ziyu Guo, Rain Liu, Xinyan Chen +1

Meta AI · The Chinese University of Hong Kong

Visual Reasoning Large Vision-Language Models Latent Visual Reasoning VLM Reasoning RL for VLMs

May 14, 2026cs.CV

MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs

Large vision-language models (LVLMs) have achieved remarkable performance across diverse multimodal tasks, yet they continue to suffer from hallucinations, generating content that is inconsistent with the visual input. Prior work DHCP (Detecting Hallucinations by Cross-modal Attention Pattern) has explored hallucination detection from the perspective of cross-modal attention, but does not address hallucination mitigation. In this paper, we propose MHSA (Mitigating Hallucinations via Steered Attention), a lightweight framework that mitigates hallucinations by learning to correct cross-modal attention patterns in LVLMs. MHSA trains a simple three-layer MLP generator to produce corrected attention, guided by supervisory signals from the DHCP discriminator and the LVLM itself. During inference, MHSA mitigates both discriminative and generative hallucinations across various datasets and LVLMs by simply replacing the original cross-modal attention with the corrected one, without modifying any LVLM parameters. By extending cross-modal attention mechanisms from hallucination detection to hallucination mitigation, MHSA offers a novel perspective on hallucination research in LVLMs and helps enhance their reliability.

Wei Ding, Yilin Li, Yudong Zhang +4

Tsinghua University Beijing, China · Tsinghua University, Tencent Beijing, China · Tencent Beijing, China +2

Vision-Language Models LLM Hallucination Mitigation Hallucination in Language Models VLM Hallucination Large Vision-Language Models Cross-Modal Attention+1

May 14, 2026cs.CV

Do We Really Need External Tools to Mitigate Hallucinations? SIRA: Shared-Prefix Internal Reconstruction of Attribution

Large vision-language models (LVLMs) often hallucinate when language priors dominate weak or ambiguous visual evidence. Existing contrastive decoding methods mitigate this problem by comparing predictions from the original image with those from externally perturbed visual inputs, but such references can introduce off-manifold artifacts and require costly extra forward passes. We propose SIRA, a training-free internal contrastive decoding framework that constructs a counterfactual reference inside the same LVLM by exploiting the staged information flow of multimodal transformers. Instead of removing visual information from the input, SIRA first lets image and text tokens interact through a shared prefix, forming an aligned multimodal state that preserves prompt interpretation, decoding history, positional structure, and early visual grounding. It then forks a counterfactual branch in later transformer layers, where attention to image-token positions is masked. This branch retains the shared multimodal context but lacks continued access to fine-grained visual evidence, yielding a language-prior-dominated internal reference for token-level contrast. During decoding, SIRA suppresses tokens that remain strong without late visual access and favors predictions whose advantage depends on the full visual pathway. Experiments on POPE, CHAIR, and AMBER with Qwen2.5-VL and LLaVA-v1.5 show that SIRA consistently reduces hallucinations while preserving descriptive coverage and incurring lower overhead than two-pass contrastive decoding. SIRA requires no training, external verifier, or perturbed input, and applies to open-weight LVLMs with white-box inference access.

Tian Qin, Junzhe Chen, Yuqing Shi +3

Tsinghua University · The University of Sydney · Stanford University +1

Efficient VLM Inference Hallucination in Language Models VLM Hallucination Large Vision-Language Models Multimodal Foundation Models Contrastive Decoding+2

May 14, 2026cs.CV

CoRDS: Coreset-based Representative and Diverse Selection for Streaming Video Understanding

Streaming video understanding with large vision-language models (VLMs) requires a compact memory that can support future reasoning over an ever-growing visual history. A common solution is to compress the key-value (KV) cache, but existing streaming methods typically rely on local token-wise heuristics, such as recency, temporal redundancy, or saliency, which do not explicitly optimize whether the retained cache is representative of the accumulated history. We propose to view KV-cache compression as a coreset selection problem: rather than scoring tokens independently for retention, we select a small subset that covers the geometry of the accumulated visual cache. Our method operates in a joint KV representation and introduces a bicriteria objective that balances coverage in key and value spaces, preserving both retrieval structure and output-relevant information. To encourage a more diverse retained subset, we further introduce an orthogonality-driven diversity criterion that favors candidates contributing new directions beyond the current selection, and connect this criterion to log-determinant subset selection. Across four open-source VLMs and five long-video and streaming-video benchmarks, our method improves over heuristic streaming compression baselines under a fixed cache budget. These results highlight that representative coreset selection offers a more effective principle, than token-wise pruning, for memory-constrained streaming video understanding.

Ailar Mahdizadeh, Puria Azadi, Muchen Li +2

University of British Columbia · Vector Institute

Streaming Video Understanding KV Caching Large Vision-Language Models Coreset Selection KV-Cache Compression Feature Selection

May 14, 2026cs.CR

To See is Not to Learn: Protecting Multimodal Data from Unauthorized Fine-Tuning of Large Vision-Language Model

The rapid advancement of Large Vision-Language Models (LVLMs) is increasingly accompanied by unauthorized scraping and training on multimodal web data, posing severe copyright and privacy risks to data owners. Existing countermeasures, such as machine unlearning and watermarks, are inherent post-hoc approaches that act only after intellectual property infringement has already occurred. In this work, we propose MMGuard to empower data owners to proactively protect their multimodal data against unauthorized LVLM fine-tuning. MMGuard generates unlearnable examples by injecting human-imperceptible perturbations that actively exploit the learning dynamics of LVLMs. By minimizing the training loss, the perturbation creates an optimization shortcut, causing the model to overfit to the noise and thereby degrading downstream performance when the perturbation is absent during inference. To further strengthen this defense, MMGuard introduces a cross-modal binding disruption, strategically shifting LVLM attention to enforce a spurious correlation between the noise and the training target with theoretical guarantees. Enhanced by an ensemble learning strategy for cross-model transferability, MMGuard is evaluated against nine open-source LVLMs across six datasets. Our comprehensive results demonstrate effective, stealthy, and robust protection under white-box, gray-box, and black-box threat models, establishing a mechanistic advantage in proactively defending against aggressive fine-tuning exploitation.

Chengshuai Zhao, Zhen Tan, Dawei Li +2

School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ, USA · Department of Computer Science and Engineering, Texas A&M University, College Station, TX, USA.

VLM Robustness Multimodal Robustness Large Vision-Language Models Adversarial Examples Unlearnable Examples

May 13, 2026cs.CV

Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context

Long-context modeling is becoming a core capability of modern large vision-language models (LVLMs), enabling sustained context management across long-document understanding, video analysis, and multi-turn tool use in agentic workflows. Yet practical training recipes remain insufficiently explored, particularly for designing and balancing long-context data mixtures. In this work, we present a systematic study of long-context continued pre-training for LVLMs, extending a 7B model from 32K to 128K context with extensive ablations on long-document data. We first show that long-document VQA is substantially more effective than OCR transcription. Building on this observation, our ablations further yield three key findings: i) for sequence-length distribution, balanced data outperforms target-length-focused data (e.g., 128K), suggesting that long-context ability requires generalizable key-information retrieval across various lengths and positions; ii) retrieval remains the primary bottleneck, favoring retrieval-heavy mixtures with modest reasoning data for task diversity; and iii) pure long-document VQA largely preserves short-context capabilities, suggesting that instruction-formatted long data reduces the need for short-data mixing. Based on these findings, we introduce MMProLong, obtained by long-context continued pre-training from Qwen2.5-VL-7B with only a 5B-token budget. MMProLong improves long-document VQA scores by 7.1% and maintains strong performance at 256K and 512K contexts beyond its 128K training window, without additional training. It further generalizes to webpage-based multimodal needle retrieval, long-context vision-text compression, and long-video understanding without task-specific supervision. Overall, our study establishes a practical LongPT recipe and an empirical foundation for advancing long-context vision-language models.

Zhaowei Wang, Lishu Luo, Haodong Duan +9

1CSE Department, HKUST · 2ByteDance Seed · *Work done at ByteDance Seed

Visual Question Answering Vision-Language Models Large Vision-Language Models Long-Context Modeling Vision-Language Pretraining Long-Context VLMs

May 12, 2026cs.CV

Is Video Anomaly Detection Misframed? Evidence from LLM-Based and Multi-Scene Models

Recent video anomaly detection research has expanded rapidly with an emphasis on general models of normality intended to work across many different scenes. While this focus has led to improvements in scalability and multi-scene generalization, it has also shifted the field away from modeling the scene-specific and context-dependent nature of normal behavior. Contemporary approaches frequently rely on video-level weak supervision and opaque pretrained representations from multi-modal large language models (MLLMs), which encourage models to respond to familiar semantic anomaly categories rather than to deviations from the normal patterns of a particular environment. This trend suppresses spatial localization, introduces semantic bias, and reduces anomaly detection to a form of action recognition. In this paper, we examine whether these prevailing formulations align with the core requirements of real-world VAD, which is typically performed within a single scene where normality is determined by local geometry, semantics, and activity patterns. Through targeted visual analyses and empirical evaluations, we demonstrate the practical consequences of these limitations and show that meaningful progress in VAD requires renewed focus on single-scene, spatially-aware, and explainable formulations that capture the nuanced structure of normality within individual environments.

Furkan Mumcu, Michael J. Jones, Anoop Cherian +1

University of South Florida · Mitsubishi Electric Research Laboratories (MERL)

Large Vision-Language Models Anomaly Localization Video Anomaly Detection Weakly Supervised Video Anomaly Detection

May 12, 2026cs.LG

Instruction Lens Score: Your Instruction Contributes a Powerful Object Hallucination Detector for Multimodal Large Language Models

Multimodal large language models (MLLMs) have achieved remarkable progress, yet the object hallucination remains a critical challenge for reliable deployment. In this paper, we present an in-depth analysis of instruction token embeddings and reveal that they implicitly encode visual information while effectively filtering erroneous information introduced by misleading visual embeddings. Building on this insight, we propose the Instruction Lens Score (InsLen), which combines a Calibrated Local Score with a Context Consistency Score that measures context consistency of the object tokens. The proposed approach serves as a plug-and-play object hallucination detector without relying on auxiliary models or additional training. Extensive experiments across multiple benchmarks and diverse MLLM architectures demonstrate that InsLen consistently outperforms existing hallucination detection methods, highlighting its effectiveness and robustness. The code is available at https://github.com/Fraserlairh/Instruction-Lens-Score.

Runhe Lai, Xinhua Lu, Yanqi Wu +3

School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China · Peng Cheng Laboratory, Shenzhen, China · Key Laboratory of Machine Intelligence and Advanced Computing, MOE, Guangzhou, China.

Hallucination Detection in VLMs Hallucination Detection Hallucination in Language Models Large Vision-Language Models Object Hallucination in VLMs LLM Hallucination Detection+1

May 12, 2026cs.CV

VideoSEAL: Mitigating Evidence Misalignment in Agentic Long Video Understanding by Decoupling Answer Authority

Long video question answering requires locating sparse, time-scattered visual evidence within highly redundant content. Although current MLLMs perform well on short videos, long videos introduce long-horizon search and verification, which often necessitates multi-turn, agentic interaction. We show that existing LVU agents can exhibit "evidence misalignment": they produce correct answers that are not supported by the retrieved or inspected evidence. To characterize this failure, we introduce two diagnostics (temporal groundedness and semantic groundedness) and use them to reveal two pressures that amplify misalignment: prompt pressure from shared-context saturation at inference time and reward pressure from outcome-only optimization during training. These findings point to a structural root cause: the coupled agent paradigm conflates long-horizon planning with answer authority. We therefore propose the decoupled planner-inspector framework, which separates planning from answer authority and gates final answering on pixel-level verification. Across four long-video benchmarks, our framework improves both answer accuracy and evidence alignment, achieving 55.1% on LVBench and 62.0% on LongVideoBench while producing interpretable search trajectories. Moreover, the decoupled architecture scales consistently with increased search budgets and supports plug-and-play upgrades of the MLLM backbone without retraining the planner. Code and models are available at https://github.com/Echochef/VideoSEAL.

Chenhao Qiu, Yechao Zhang, Xin Luo +2

Mango TV · Mango TV, Changsha, China · Nanyang Technological University +1

Large Vision-Language Models Video QA Tool-Using Vision-Language Agents Long-Video QA