视觉-语言模型中提示诱导幻觉的机制
Mechanisms of Prompt-Induced Hallucination in Vision-Language Models
AI总结：
研究揭示了视觉-语言模型在提示诱导幻觉中的机制，通过分析三种模型发现特定注意力头可显著减少幻觉现象，揭示了模型对提示的响应差异。
摘要：
大型视觉-语言模型（VLMs）虽然能力强大，但常因优先文本提示而产生幻觉。本文在受控的物体计数设置中研究这一失效模式，发现当物体数量增加时，模型更倾向于遵循提示而非视觉证据。通过分析三种VLMs，识别出一组注意力头，其移除可使提示诱导幻觉（PIH）减少至少40%。不同模型中，PIH头以模型特定的方式介导提示复制，研究显示移除PIH头可增加对视觉证据的修正。本文揭示了驱动提示诱导幻觉的内部机制，展示了模型在实现这些行为上的差异。
英文摘要：
Large vision-language models (VLMs) are highly capable, yet often hallucinate by favoring textual prompts over visual evidence. We study this failure mode in a controlled object-counting setting, where the prompt overstates the number of objects in the image (e.g., asking a model to describe four waterlilies when only three are present). At low object counts, models often correct the overestimation, but as the number of objects increases, they increasingly conform to the prompt regardless of the discrepancy. Through mechanistic analysis of three VLMs, we identify a small set of attention heads whose ablation substantially reduces prompt-induced hallucinations (PIH) by at least 40% without additional training. Across models, PIH-heads mediate prompt copying in model-specific ways. We characterize these differences and show that PIH ablation increases correction toward visual evidence. Our findings offer insights into the internal mechanisms driving prompt-induced hallucinations, revealing model-specific differences in how these behaviors are implemented.
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。
