揭秘大模型如何识别图像中的文字，发现专用文字识别头。
How Do Large Vision-Language Models See Text in Image? Unveiling the Distinctive Role of OCR Heads
- 定位图像文字时，多个神经元头被激活，非稀疏分布。
- 文字识别头特征与普通检索头差异显著，独立运作。
- 适合研究模型可解释性或提升图文理解性能的开发者。
尽管大型视觉语言模型（LVLMs）取得显著进展，但在可解释性以及如何定位和解析图像中文字信息方面仍存在差距。本文研究多种LVLMs，识别出负责从图像中提取文字信息的特定神经头，称为光学字符识别头（OCR Head）。主要发现包括：(1) 非稀疏性：与以往检索头不同，大量头部被激活以提取图像中的文本信息；(2) 特征迥异：OCR头在性质上显著区别于通用检索头，特征相似度低；(3) 静态激活：这些头部的激活频率与其OCR得分高度一致。通过链式思维（CoT）和头部屏蔽实验验证了上述结论，并证明重新分配OCR头中的sink-token值可提升性能。研究深化了对LVLM处理图像嵌入文本机制的理解。
原文摘要 · Abstract (English)
Despite significant advancements in Large Vision Language Models (LVLMs), a gap remains, particularly regarding their interpretability and how they locate and interpret textual information within images. In this paper, we explore various LVLMs to identify the specific heads responsible for recognizing text from images, which we term the Optical Character Recognition Head (OCR Head). Our findings regarding these heads are as follows: (1) Less Sparse: Unlike previous retrieval heads, a large number of heads are activated to extract textual information from images. (2) Qualitatively Distinct: OCR heads possess properties that differ significantly from general retrieval heads, exhibiting low similarity in their characteristics. (3) Statically Activated: The frequency of activation for these heads closely aligns with their OCR scores. We validate our findings in downstream tasks by applying Chain-of-Thought (CoT) to both OCR and conventional retrieval heads and by masking these heads. We also demonstrate that redistributing sink-token values within the OCR heads improves performance. These insights provide a deeper understanding of the internal mechanisms LVLMs employ in processing embedded textual information in images.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。