paper-with-me

Papers

Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs

2024-11-15 · Xiaofeng Zhang, Yihao Quan, Chaochen Gu, Chen Shen, Xiaosong Yuan, Shaotian Yan, Hao Cheng, Kaijie Wu, Jieping Ye

The hallucination problem in multimodal large language models (MLLMs) remains a common issue. Although image tokens occupy a majority of the input sequence of MLLMs, there is limited research to explore the relationship between image tokens and hallucinations. In this paper, we analyze the distribution of attention scores for image tokens across each layer and head of the model, revealing an intriguing and common phenomenon: most hallucinations are closely linked to the pattern of attention sinks in the self-attention matrix of image tokens, where shallow layers exhibit dense attention sinks and deeper layers show sparse attention sinks. We further analyze the attention heads of different layers and find that heads with high-density attention sink in the image part play a positive role in alleviating hallucinations. In this paper, we propose a training-free method named \textcolor{red}{\textbf{E}}nhancing \textcolor{red}{\textbf{A}}ttention \textcolor{red}{\textbf{H}}eads (EAH), an approach designed to enhance the convergence of image tokens attention sinks in the shallow layers. EAH identifies the attention head that shows the vision sink in a shallow layer and extracts its attention matrix. This attention map is then broadcast to other heads in the layer, thereby strengthening the layer to pay more attention to the image itself. With extensive experiments, EAH shows significant hallucination-mitigating performance on different MLLMs and metrics, proving its effectiveness and generality.

📄 PDF Abstract BibTeX arXiv:2411.09968

Code (0)

등록된 구현이 없습니다.

Tasks

Hallucination

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음
Attention Sinks Please enter a description about the method here

Similar Papers 제목 키워드 기반

Seeing Right but Saying Wrong: Inter- and Intra-Layer Refinement in MLLMs without Training

2026-01-12 · Shezheng Song, Shasha Li, Jie Yu arxiv

Multimodal Large Language Models (MLLMs) have demonstrated strong capabilities across a variety of vision-language tasks. However, their internal reasoning often exhibits a critical inconsistency: although deeper layers …

On the Importance of Local Information in Transformer Based Models

2020-08-13 · Madhura Pande, Aakriti Budhraja, Preksha Nema, Pratyush Kumar 외

The self-attention module is a key component of Transformer-based models, wherein each token pays attention to every other token. Recent studies have shown that these heads exhibit syntactic, semantic, or local behaviour…

de-enMRPCQNLIQQP+1

When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models

2025-07-18 · Francesco Ortu, Zhijing Jin, Diego Doimo, Alberto Cazzaniga arxiv

Vision-language models (VLMs) increasingly combine visual and textual information to perform complex tasks. However, conflicts between their internal knowledge and external visual input can lead to hallucinations and unr…

Shapley Head Pruning: Identifying and Removing Interference in Multilingual Transformers

2022-10-11 · William Held, Diyi Yang

Multilingual transformer-based models demonstrate remarkable zero and few-shot transfer across languages by learning and reusing language-agnostic features. However, as a fixed-size model acquires more languages, its per…

SentenceSentence Classification

Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis

2024-09-21 · Zeping Yu, Sophia Ananiadou

We find arithmetic ability resides within a limited number of attention heads, with each head specializing in distinct operations. To delve into the reason, we introduce the Comparative Neuron Analysis (CNA) method, whic…

Model EditingPrediction