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 demonstrates that vision‑language models (VLMs) possess a small set of attention heads, called Visual Retrieval Heads (VRHs), that are causally responsible for linking text prompts to specific image regions. By adapting head‑scoring techniques from language models, the authors identify VRHs as the heads whose attention from output prediction tokens, summed over the ground‑truth referent region, most reliably indicates causal grounding. Experiments across eleven VLMs and five referring‑expression benchmarks show that masking the top 20 VRHs can drop grounding accuracy by up to 80 percentage points, while random masking has little effect, and that VRHs generalize across diverse visual tasks and transfer across models sharing an LLM backbone.
arXiv:2606. 05843v1 Announce Type: cross Abstract: While Multimodal Large Language Models (MLLMs) demonstrate remarkable proficiency on complex vision-language tasks, the mechanisms by which they extract query-relevant visual features from complex, noisy contexts remain opaque.
By Ruoxi Sun, Quantong Qiu, Juntao Li, Zecheng Tang, Yihang Lou, Min Zhang
arXiv:2606. 04240v1 Announce Type: cross Abstract: Retrieval over visually-rich documents, pages that interleave text with figures, tables, and charts, is essential for multimodal retrieval-augmented generation, yet most retrievers still discard the visual channel.
By Jingbiao Mei
arXiv:2604. 01280v2 Announce Type: replace-cross Abstract: Knowledge-based Visual Question Answering (KB-VQA) requires Multimodal Large Language Models (MLLMs) to identify and combine fine-grained visual cues with retrieved textual evidence.
By Marco Morini, Sara Sarto, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara
arXiv:2606. 16494v1 Announce Type: cross Abstract: Knowledge-based visual question answering (KB-VQA) lets vision-language systems answer questions that exceed their parametric knowledge by conditioning a reader on passages retrieved from a Wikipedia-scale knowledge base.
By Jieyuan Liu, Jianyang Gu, Shijie Chen, Jefferson Chen, Zhen Wang
arXiv:2607. 24748v1 Announce Type: cross Abstract: Visually-rich documents such as reports, slides, and manuals often distribute the evidence needed to answer a question across multiple pages, mixing text with layout cues, tables, charts, and figures.
By Seonok Kim
ET‑Prune is a training‑free framework that dynamically allocates visual token budgets in multimodal large language models based on question‑conditioned evidence. It protects text‑like spatial regions, converts evidence uncertainty into a token floor, and progressively prunes concentrated evidence while retaining diffuse or text‑dense tokens. In six backbone‑benchmark comparisons, ET‑Prune matches or outperforms other pruned methods while keeping roughly half the visual tokens, achieving notable gains on OCRBench‑v2 and MMBench v1.1.
By Zizhong Ding, Junxian Li, Kai Liu, Shaoqiu Zhang, Xiao Xiao, Linghe Kong, Yulun Zhang
MIDR (Multimodal Indexing for Document Retrieval) is a training‑free framework that enriches document indexes by converting rendered pages into verified textual fields with a multimodal LLM, then indexing those fields with BM25F and optionally fusing with dense retrieval. By shifting multimodal reasoning to index time, MIDR enables text‑centric serving while retaining multimodal evidence, achieving a 23.0% relative gain over BM25 on ViDoRe V3 and outperforming ColQwen2.5 on several domains with significantly smaller index memory and lower query latency.
By Debanjan Mahata, Atharva Tendle, Daniel Preotiuc-Pietro, Yong Zhuang, Ozan Irsoy
PailitaoGR is a generative image retrieval model that incorporates a latent think-with-images approach to better handle real‑world query images. It uses a target‑focused perception mechanism—comprising a target enhancer and on‑policy distillation—to highlight the search target, and a selective auxiliary‑evidence mechanism—using an auxiliary enhancer and incremental contrastive distillation—to exploit useful side information. Trained on real‑world online image‑search logs, the method achieves an average 13.8 % improvement over existing baselines.
By Xiaomeng Fan, Yueran Liu, Shengyu Zhou, Chenghan Fu, Wanxian Guan, Feng Li, Chuan Yu, Jian Xu, Bo Zheng
arXiv:2602. 00344v2 Announce Type: replace-cross Abstract: While Retrieval-Augmented Generation (RAG) is one of the dominant paradigms for enhancing Large Vision-Language Models (LVLMs) on knowledge-based VQA tasks, recent work attributes RAG failures to insufficient attention towards the retrieved context, proposing to reduce the attention allocated to image tokens.
By Beidi Zhao, Wenlong Deng, Xinting Liao, Yushu Li, Nazim Shaikh, Yao Nie, Xiaoxiao Li
External memory effectively grounds large language models (LLMs) and vision-language models (VLMs)-based question answering (QA) in relevant multimodal evidence. However, existing memory paradigms represent each memory item in raw text and image forms, so retrieval-based systems must pass the retrieved text or images to the generation LLMs/VLMs, resulting in high token consumption and storage pressure, making it unaffordable for resource-constrained applications.