Aligning Attention Distribution to Information Flow for Hallucination Mitigation in Large Vision-Language Models

• URL: http://arxiv.org/abs/2505.14257v1
• Date: Tue, 20 May 2025 12:10:13 GMT
• Title: Aligning Attention Distribution to Information Flow for Hallucination Mitigation in Large Vision-Language Models
• Authors: Jianfei Zhao, Feng Zhang, Xin Sun, Chong Feng,
• Abstract summary: We enhance the model's visual understanding by leveraging the core information embedded in semantic representations.<n>We evaluate our method on three image captioning benchmarks using five different LVLMs, demonstrating its effectiveness in significantly reducing hallucinations.
• Score: 11.385588803559733
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Due to the unidirectional masking mechanism, Decoder-Only models propagate information from left to right. LVLMs (Large Vision-Language Models) follow the same architecture, with visual information gradually integrated into semantic representations during forward propagation. Through systematic analysis, we observe that over 80\% of the visual information is absorbed into the semantic representations. However, the model's attention still predominantly focuses on the visual representations. This misalignment between the attention distribution and the actual information flow undermines the model's visual understanding ability and contributes to hallucinations. To address this issue, we enhance the model's visual understanding by leveraging the core information embedded in semantic representations. Specifically, we identify attention heads that focus on core semantic representations based on their attention distributions. Then, through a two-stage optimization paradigm, we propagate the advantages of these attention heads across the entire model, aligning the attention distribution with the actual information flow. We evaluate our method on three image captioning benchmarks using five different LVLMs, demonstrating its effectiveness in significantly reducing hallucinations. Further experiments reveal a trade-off between reduced hallucinations and richer details. Notably, our method allows for manual adjustment of the model's conservativeness, enabling flexible control to meet diverse real-world requirements. Code will be released once accepted.

Related papers

• CAI: Caption-Sensitive Attention Intervention for Mitigating Object Hallucination in Large Vision-Language Models [60.0300765815417]
  Large Vision-Language Models (LVLMs) frequently produce content that deviates from visual information, leading to object hallucination.<n>We propose Caption-sensitive Attention Intervention (CAI), a training-free, plug-and-play hallucination mitigation method.
  arXiv  Detail & Related papers  (2025-06-30T07:52:36Z)
• GLIMPSE: Holistic Cross-Modal Explainability for Large Vision-Language Models [0.0]
  We introduce GLIMPSE, a model-agnostic framework that jointly attributes LVLM outputs to the most relevant visual evidence and textual signals.<n>GLIMPSE fuses gradient-weighted attention, adaptive layer propagation, and relevance-weighted token aggregation to produce holistic response-level heat maps.<n>We demonstrate an analytic approach to uncover fine-grained insights into LVLM cross-modal attribution, trace reasoning dynamics, analyze systematic misalignment, diagnose hallucination and bias, and ensure transparency.
  arXiv  Detail & Related papers  (2025-06-23T18:00:04Z)
• SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding [5.976839106353883]
  SECOND: Selective and Contrastive Decoding is a novel approach that enables Vision-Language Models to leverage multi-scale visual information with an object-centric manner.<n> SECOND significantly reduces perceptual hallucinations and outperforms a wide range of benchmarks.
  arXiv  Detail & Related papers  (2025-06-10T02:55:38Z)
• Exploring Implicit Visual Misunderstandings in Multimodal Large Language Models through Attention Analysis [21.869968563545736]
  We define implicit visual misunderstanding (IVM), where MLLMs provide correct answers without fully comprehending the visual input.<n>We introduce a scale-agnostic metric, textitattention accuracy, and a novel benchmark for quantifying IVMs.<n>We extend our approach to finer granularities and demonstrate its effectiveness in unimodal scenarios.
  arXiv  Detail & Related papers  (2025-05-15T17:52:40Z)
• MINT: Mitigating Hallucinations in Large Vision-Language Models via Token Reduction [6.416957959150438]
  Hallucinations hinder the application of Large Vision-Language Models (LVLMs) in domains that require high reliability.<n>We propose MINT, a training-free decoding strategy, MItigating hallucinations via tokeN reducTion.<n>Our approach achieves a 4% improvement in mitigating hallucinations caused by distracted perception compared to original models.
  arXiv  Detail & Related papers  (2025-02-02T08:34:57Z)
• PAINT: Paying Attention to INformed Tokens to Mitigate Hallucination in Large Vision-Language Model [0.0]
  hallucinations often arise from the progressive weakening of attention weight to visual tokens.<n>textbfPAINT (textbfPaying textbfAttention to textbfINformed textbfTokens) is a plug-and-play framework that intervenes in the self-attention mechanism of the Large Vision Language Models.
  arXiv  Detail & Related papers  (2025-01-21T15:22:31Z)
• Mitigating Hallucination for Large Vision Language Model by Inter-Modality Correlation Calibration Decoding [66.06337890279839]
  Large vision-language models (LVLMs) have shown remarkable capabilities in visual-language understanding for downstream multi-modal tasks.<n>LVLMs still suffer from generating hallucinations in complex generation tasks, leading to inconsistencies between visual inputs and generated content.<n>We propose an Inter-Modality Correlation Decoding (IMCCD) method to mitigate hallucinations in LVLMs in a training-free manner.
  arXiv  Detail & Related papers  (2025-01-03T17:56:28Z)
• ICT: Image-Object Cross-Level Trusted Intervention for Mitigating Object Hallucination in Large Vision-Language Models [32.24716280370563]
  ICT is a lightweight, training-free method that calculates an intervention direction to shift the model's focus towards different levels of visual information. It achieves strong performance with a small amount of data and generalizes well across different datasets and models.
  arXiv  Detail & Related papers  (2024-11-22T12:22:21Z)
• Human-Object Interaction Detection Collaborated with Large Relation-driven Diffusion Models [65.82564074712836]
  We introduce DIFfusionHOI, a new HOI detector shedding light on text-to-image diffusion models. We first devise an inversion-based strategy to learn the expression of relation patterns between humans and objects in embedding space. These learned relation embeddings then serve as textual prompts, to steer diffusion models generate images that depict specific interactions.
  arXiv  Detail & Related papers  (2024-10-26T12:00:33Z)
• Diffusion Model with Cross Attention as an Inductive Bias for Disentanglement [58.9768112704998]
  Disentangled representation learning strives to extract the intrinsic factors within observed data. We introduce a new perspective and framework, demonstrating that diffusion models with cross-attention can serve as a powerful inductive bias. This is the first work to reveal the potent disentanglement capability of diffusion models with cross-attention, requiring no complex designs.
  arXiv  Detail & Related papers  (2024-02-15T05:07:54Z)
• Harnessing Diffusion Models for Visual Perception with Meta Prompts [68.78938846041767]
  We propose a simple yet effective scheme to harness a diffusion model for visual perception tasks. We introduce learnable embeddings (meta prompts) to the pre-trained diffusion models to extract proper features for perception. Our approach achieves new performance records in depth estimation tasks on NYU depth V2 and KITTI, and in semantic segmentation task on CityScapes.
  arXiv  Detail & Related papers  (2023-12-22T14:40:55Z)
• Unleashing Text-to-Image Diffusion Models for Visual Perception [84.41514649568094]
  VPD (Visual Perception with a pre-trained diffusion model) is a new framework that exploits the semantic information of a pre-trained text-to-image diffusion model in visual perception tasks. We show that VPD can be faster adapted to downstream visual perception tasks using the proposed VPD.
  arXiv  Detail & Related papers  (2023-03-03T18:59:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.