arXiv Computer Vision

State-Conditioned Visual Evidence Retrieval for Fine-Grained Perception in Document Vision-Language Models

arXiv Computer Vision
Aug 28

Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information

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
Hugging Face Trending Papers
Aug 27

Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information

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 AI
Sep 3

RVSD: Retrieval Vision Sparse Decoding for Mitigating Visual Hallucinations in Large Vision-Language Models

The paper introduces RVSD, a training‑free, plug‑and‑play decoding framework that combines token sparsification with Semantic‑Space Visual Retrieval (SSVR) to reduce visual hallucinations in large vision‑language models. RVSD employs a semantics‑directed token selection strategy to prune redundant tokens while preserving essential visual information, and uses SSVR to perform on‑demand cross‑modal retrieval within a shared semantic space. Experiments show that RVSD achieves state‑of‑the‑art performance in mitigating visual hallucinations while maintaining strong suppression in long‑context generation.

By Canjie Liu, Jiawen Kang, Jinbo Wen, Zishao Zhong
arXiv Computer Vision
Aug 25

FOVEA: Focused On-Demand Visual Evidence Adaptation for Cache-Friendly Multimodal Speculative Decoding

FOVEA introduces a cache‑friendly, on‑demand visual evidence adaptation for multimodal speculative decoding, enabling a lightweight draft model to dynamically retrieve a bounded subset of visual memory based on a cumulative‑mass rule. The retrieved visual readout is fused with the draft hidden state via a lightweight gated residual correction, avoiding the insertion of visual tokens into the autoregressive context. Experiments on various vision‑language backbones and benchmarks show that FOVEA improves draft acceptance and speeds up end‑to‑end decoding by up to 2.13× compared to traditional autoregressive decoding.

By Hengjie Zhu, Dayan Wu, Zihao Zhang, Xinze Liu, Jingxuan Yu, Peng Fu, Zheng Lin, Weiping Wang, Ding Wang
Hugging Face Trending Papers
Jul 27

MAViE: A Multi-scale Adaptive Vision Encoder for Fine-grained Visual Perception and Efficient Multimodal Reasoning

Vision-language models commonly project all tokens produced by a pretrained vision encoder into a large language model. However, final-layer features can discard text, local attributes, and spatial relationships, while high-resolution inputs substantially increase context length and inference latency.