arXiv AI

Visuals Lie, Consistency Speaks: Disentangling Spatial Attention from Reliability in Vision-Language Models

arXiv:2606. 17389v1 Announce Type: cross Abstract: Multimodal Foundation Models are increasingly used as reasoning agents, making reliability, knowing when a model may hallucinate, critical.

arXiv AI

Sep 2

Visual Attention Faithfulness in Vision-Language Models is Heterogeneous

The study investigates whether attention weights in Vision‑Language Models (VLMs) accurately reflect model reasoning for visual inputs. Using causal perturbation analysis, it identifies three distinct processing modes—Faithful‑Sufficient, Faithful‑Distributed, and Non‑Focal—indicating heterogeneous visual attention faithfulness. The research also shows that human‑annotated ground‑truth regions align with model attention in only about 60% of cases, highlighting a systematic divergence between model visual reliance and human intuition across VQA, document, and chart tasks.

By Xurui Song, Weishi Wang, Zhongqi Yue, Kuluhan Binici, Tao Bai, Hongxin Shao, Daniel Dahlmeier, Jun Luo

llms nlp multimodal

arXiv AI

Sep 16

Same Answer, Different Representations: Hidden instability in VLMs

arXiv:2602.06652v2 Announce Type: replace Abstract: The robustness of Vision Language Models (VLMs) is commonly assessed through output-level invariance, implicitly assuming that stable predictions r...

By Farooq Ahmad Wani, Alessandro Suglia, Rohit Saxena, Aryo Pradipta Gema, Wai-Chung Kwan, Fazl Barez, Maria Sofia Bucarelli, Fabrizio Silvestri, Pasquale Minervini

llms rag multimodal benchmarks safety

arXiv AI

Aug 3

Step-Level Visual Grounding Faithfulness Predicts Out-of-Distribution Generalization in Long-Horizon Vision-Language Models

arXiv:2603. 06828v2 Announce Type: replace-cross Abstract: We uncover a behavioral law of long-horizon vision-language models: models that maintain temporally grounded beliefs generalize better.

By Md Ashikur Rahman, Md Arifur Rahman, Niamul Hassan Samin, Abdullah Ibne Hanif Arean, Juena Ahmed Noshin

llms multimodal benchmarks

arXiv Computer Vision

Sep 16

What Do Hallucinations Reveal About Multimodal Reasoning? Diagnosing Visual Grounding Failures via Contrastive Decoding Probes

arXiv:2609.16646v1 Announce Type: new Abstract: When strong multimodal models are widely available, progress requires new scientific methodologies beyond benchmark scores---using models as instrument...

By Zhipeng Zhao, Wenxu Wang, Peishun Liu, Ruichun Tang

llms multimodal benchmarks safety

arXiv AI

Jul 28

DICA: Dual-Indicator Guided Contrastive Alignment in Multimodal Large Language Models

arXiv:2607. 23944v1 Announce Type: new Abstract: Human visual reasoning typically follows a coarse-to-fine attention process, starting from global scene understanding and gradually focusing on question-relevant regions.

By Hao Yang, Jin Wang, Xuejie Zhang

llms multimodal benchmarks safety

arXiv AI

Aug 6

Reasoning Dynamics and the Limits of Monitoring Modality Reliance in Vision-Language Models

arXiv:2604. 14888v3 Announce Type: replace-cross Abstract: Recent advances in vision language models (VLMs) offer reasoning capabilities, yet how these unfold and integrate visual and textual information remains unclear.

By Danae S\'anchez Villegas, Samuel Lewis-Lim, Nikolaos Aletras, Desmond Elliott

llms multimodal safety

arXiv AI

Jun 8

The Geometry of Representational Failures in Vision Language Models

arXiv:2602. 07025v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) exhibit puzzling failures in multi-object visual tasks, such as hallucinating non-existent elements or failing to identify the most similar objects among distractions.

By Daniele Savietto, Declan Campbell, Andr\'e Panisson, Marco Nurisso, Giovanni Petri, Jonathan D. Cohen, Alan Perotti

llms multimodal

arXiv AI

Sep 10

Emergent Goal-Directed Attention in Large Vision-Language Models

arXiv:2609.05517v1 Announce Type: cross Abstract: Human observers prioritize visual information according to task goals. Most computational models of naturalistic viewing are gaze-trained for free vi...

By Han Zhang

llms multimodal safety

arXiv AI

Jun 29

Dismantling Pathological Shortcuts: A Causal Framework for Faithful LVLM Decoding

arXiv:2606. 27596v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) exhibit sophisticated reasoning but remain susceptible to object hallucination.

By Liu Yu, Can Chen, Ping Kuang, Zhikun Feng, Fan Zhou, Gillian Dobbie

llms multimodal benchmarks safety

arXiv Computer Vision

Sep 22

Seeing is not Enough: Vision-Language Models Perceive Evidence but Fail to Act

arXiv:2609.22588v1 Announce Type: new Abstract: Vision-language models (VLMs) perform strongly on visual question answering benchmarks, yet often make decisions that contradict visual evidence they h...

By Yuyang Dai, Bofei Huang, Hongbo Zhang, Haoran Xie

llms nlp multimodal benchmarks

arXiv AI

Sep 18

Cross-Modal Attention Acts as a Frequency Filter: Why Verbose Prompts Improve Robustness in Vision-Language Models

The paper shows that the wording of prompts in vision‑language models (VLMs) can either improve or worsen robustness to image corruption. Verbose prompts broaden the cross‑modal attention’s frequency filter, making the model less sensitive to corruptions, while semantically complex prompts narrow the filter and increase vulnerability. Experiments on Qwen3‑VL and LLaVA‑OneVision confirm that adding padding or verbose phrasing reduces answer drift by 70–81% on 8B models.

By Farooq Ahmad Wani, Maria Sofia Bucarelli, Mujtaba Hussain Mirza, Oleksandr Pryymak, Aryo Pradipta Gema, Iacopo Masi, Pasquale Minervini, Fabrizio Silvestri

llms multimodal

arXiv Computer Vision

Aug 27

Seeing or Knowing? Visual Context Sensitivity in Multimodal Large Language Models

The paper investigates why multimodal large language models (MLLMs) struggle with vision‑centric tasks when visual evidence conflicts with pretrained language knowledge. Using image reconstruction and a new WhatIfVis benchmark, the authors show that MLLMs preserve coarse‑grained visual attributes but fail to consistently use them, and that supervised fine‑tuning and activation patching can improve controllability of visual context sensitivity. The study demonstrates that the main bottleneck lies in the models’ inability to reliably regulate their reliance on visual evidence rather than in visual perception itself.

By Jiaang Li, Chengzu Li, Zhaochong An, Yifei Yuan, Xi Liu, Serge Belongie, V\'esteinn Sn{\ae}bjarnarson

llms fine-tuning multimodal benchmarks

Related stories