Hallucinations in Large Vision-Language Models (LVLMs) significantly undermine their reliability, motivating researchers to explore the causes of hallucination. However, most studies primarily focus on the language aspect rather than the visual. In this paper, we address how LVLMs process visual information and whether this process causes hallucination. Firstly, we use the attention lens to identify the stages at which LVLMs handle visual data, discovering that the middle layers are crucial. Moreover, we find that these layers can be further divided into two stages: ''visual information enrichment'' and ''semantic refinement'' which respectively propagate visual data to object tokens and interpret it through text. By analyzing attention patterns during the visual information enrichment stage, we find that real tokens consistently receive higher attention weights than hallucinated ones, serving as a strong indicator of hallucination. Further examination of multi-head attention maps reveals that hallucination tokens often result from heads interacting with inconsistent objects. Based on these insights, we propose a simple inference-time method that adjusts visual attention by integrating information across various heads. Extensive experiments demonstrate that this approach effectively mitigates hallucinations in mainstream LVLMs without additional training costs. Code is available at https://github.com/ZhangqiJiang07/middle_layers_indicating_hallucinations.

翻译：大型视觉语言模型（LVLMs）中的幻觉现象严重削弱了其可靠性，促使研究人员深入探索幻觉的成因。然而，现有研究大多聚焦于语言层面，而忽视了视觉处理过程。本文旨在探究LVLMs如何处理视觉信息，以及这一过程是否会导致幻觉产生。首先，我们利用注意力透镜识别出LVLMs处理视觉数据的关键阶段，发现中间层在其中起着决定性作用。进一步研究发现，这些中间层可细分为两个阶段：“视觉信息增强”阶段与“语义精炼”阶段，前者负责将视觉数据传播至物体标记，后者则通过文本对信息进行解读。通过分析视觉信息增强阶段的注意力模式，我们发现真实标记始终比幻觉标记获得更高的注意力权重，这为幻觉检测提供了强有力指标。对多头注意力图的深入研究表明，幻觉标记往往源于注意力头与不一致物体的交互作用。基于这些发现，我们提出一种简单的推理时方法，通过整合多头注意力信息来调整视觉注意力权重。大量实验证明，该方法能有效缓解主流LVLMs的幻觉现象，且无需额外训练成本。代码已开源：https://github.com/ZhangqiJiang07/middle_layers_indicating_hallucinations。