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
arXiv:2605.26380v2 Announce Type: replace-cross
Abstract: Frontier multimodal large language models (MLLMs) have been reported to achieve over 90\% accuracy on fine-grained perception benchmarks. How...
By Jingru Chen, Yiming Liu, Mingtao Chen, Sijie Chen, Richeng Xuan, Liang Yang, Zhichao Hu, Fanyang Lu
arXiv:2607. 26769v1 Announce Type: cross Abstract: Multimodal large language models increasingly use sketches, annotations, tools, and intermediate images during reasoning, but it remains unclear whether they truly rely on these visual states.
By Siyu Yan, Zhuoran Yan, Haiying Xu, Panhao Zhou, Jingyu Chen, Chenhao Ji, Shuo Cao, Yongheng Zhang, Haoze Liu, Siyu Zhang, Xiwen Gu, Yihao Liu, Alex Jinpeng Wang
arXiv:2509. 22415v3 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have achieved strong vision-language performance, yet their token-level visual evidence remains difficult to inspect.
By Jiawei Liang, Jianjie Huang, Ruoyu Chen, Xianghao Jiao, Siyuan Liang, Shiming Liu, Xiaochun Cao
The study examines how Vision‑Language Models (VLMs) integrate visual evidence into language‑based decisions by applying layer‑wise causal interventions on video‑text attention pathways in a video‑based generative multiple‑choice setting. Findings reveal that visual information is primarily incorporated while processing candidate answer options, with nouns serving as key semantic anchors and verbs becoming important during temporal reasoning. The research also uncovers a distinct pattern in temporal reasoning, indicating that VLMs struggle to reconstruct sequential information across video frames, possibly due to linguistic biases in temporal expressions.
By Davide Testa, Hugh Mee Wong, Alessandro Lenci, Bernardo Magnini, Albert Gatt
The paper introduces CAIT, a benchmark of 400 synthetic scenes featuring counter‑intuitive actions that challenge multimodal large language models (MLLMs). Human participants and proprietary models like Claude and Gemini perform well, but standard open‑source instruction‑tuned MLLMs fail, largely due to a strong language prior that overrides contradictory visual evidence. The study shows that Chain‑of‑Thought reasoning can help but introduces new issues, while targeted fine‑tuning and structured prompting can reduce reliance on language priors and improve visual grounding.
By Chen Ling, Tongwei Zhang, Hanqian Li, Nai Ding
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
arXiv:2607. 03595v1 Announce Type: cross Abstract: Affordance grounding aims to localize image regions that support a specific action, serving as a core capability for physical intelligence and embodied perception.
By Seung Il Lee, Qinqian Lei, Daguang Xu, Dong Yang, Robby T. Tan, Yixin Chen, Bo Wang
arXiv:2608. 08021v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) should answer from concrete image evidence rather than language priors, dataset shortcuts, or irrelevant visual context.
By Haojie Huang, Xinlei Yu, Chengming Xu, Zhangquan Chen, Cheng Yang, Qingdong He, Yu Yang, Jiangning Zhang, Xiaobin Hu
The paper introduces Visual Retrieval Heads (VRHs), a small fraction of attention heads in vision‑language models that are causally responsible for grounding text descriptions to image regions. By recasting head‑scoring methods and evaluating across eleven VLMs and five benchmarks, the authors show that masking the top 20 VRHs can drop grounding accuracy by up to 80 percentage points, while random masking has little effect. VRHs generalize across various visual reference tasks, preserve output format while corrupting localization, and transfer causally across models sharing an LLM backbone.
By Chanho Park, Daehyeon Choi, Jihyun Lee, Minhyuk Sung
The paper introduces EviSpec, a training‑free compiler that generates complementary evidence specifications to improve high‑resolution multimodal large language models (MLLMs). By explicitly guiding visual search with structured evidence specifications, EviSpec achieves significant relative gains—up to 14.8% over random evidence—across five MLLMs and three benchmarks, and also sets new state‑of‑the‑art results on VQA and hallucination‑focused tasks.
By Zhongkuan Mao, Wenzhuo Zhao, Xianjie Liu, Yidong Wang, Zhao Gao, Ronghao Xian, Yao Jiang, Yi Zhang, Liangjian Wen, Keren Fu
arXiv:2608. 19208v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are frequently exposed to auxiliary textual context, the impact of which on visually grounded tasks remains underexplored.
By Yinfeng Wang, Zhiyuan Yao, Zheren Fu, Lei Zhang, Zhendong Mao