增强文本生成中的视觉依赖：从贝叶斯视角缓解大型视觉语言模型幻觉
Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models
AI总结：
本文从贝叶斯视角分析LVLM幻觉成因，提出移除冗余视觉token、修正先验信息以及在视觉后验坍缩时停止生成三种策略，在多个基准上有效缓解幻觉并优于现有方法。
摘要：
大型视觉语言模型（LVLMs）通常生成的文本满足上下文连贯性，但与视觉输入不匹配。这种幻觉问题阻碍了LVLMs在现实世界中的应用。解决LVLM幻觉的关键在于使文本生成更多地依赖视觉内容。以往大多数工作选择增强或调整特定模态（即视觉或文本）的特征/输出来缓解LVLM中的幻觉，但并未明确或系统地增强视觉依赖。在本文中，我们从贝叶斯视角全面研究了可能导致LVLM文本生成中视觉依赖退化的因素。基于我们的观察，我们提出从三个方面缓解LVLM中的幻觉。首先，我们观察到并非所有视觉token在生成有意义的文本时都具有信息量。我们提出评估并移除冗余的视觉token，以避免其干扰。其次，LVLM可能编码了不恰当的先验信息，使其倾向于生成意外的词语。我们提出一种简单而有效的方法，从贝叶斯视角修正先验。第三，我们观察到从某些步骤开始，以视觉token为条件的下一token预测后验可能坍缩为一个完全不依赖于任何有信息量视觉token的先验分布。因此，我们提出停止进一步的文本生成以避免幻觉。在包括POPE、CHAIR和MME在内的三个基准上的大量实验表明，我们的方法能够持续缓解LVLM的幻觉问题，并且与以往最先进的方法相比表现优异。
英文摘要：
Large Vision-Language Models (LVLMs) usually generate texts which satisfy context coherence but don't match the visual input. Such a hallucination issue hinders LVLMs' applicability in the real world. The key to solving hallucination in LVLM is to make the text generation rely more on the visual content. Most previous works choose to enhance/adjust the features/output of a specific modality (i.e., visual or textual) to alleviate hallucinations in LVLM, which do not explicitly or systematically enhance the visual reliance. In this paper, we comprehensively investigate the factors which may degenerate the visual reliance in text generation of LVLM from a Bayesian perspective. Based on our observations, we propose to mitigate hallucination in LVLM from three aspects. Firstly, we observe that not all visual tokens are informative in generating meaningful texts. We propose to evaluate and remove redundant visual tokens to avoid their disturbance. Secondly, LVLM may encode inappropriate prior information, making it lean toward generating unexpected words. We propose a simple yet effective way to rectify the prior from a Bayesian perspective. Thirdly, we observe that starting from certain steps, the posterior of next-token prediction conditioned on visual tokens may collapse to a prior distribution which does not depend on any informative visual tokens at all. Thus, we propose to stop further text generation to avoid hallucination. Extensive experiments on three benchmarks including POPE, CHAIR, and MME demonstrate that our method can consistently mitigate the hallucination issue of LVLM and performs favorably against previous state-of-the-arts.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
