发现大模型会误把图中不存在的文本当图像内容，提出检测并修正方法。
Unveiling the Response of Large Vision-Language Models to Visually Absent Tokens
- 识别出特定前馈神经元（VA神经元）能感知文本是否在图中
- 通过激活模式检测缺失视觉证据的文本，准确率超90%
- 可适配多种大模型，提升回答准确性
大型视觉语言模型（LVLMs）通过联合解析视觉与文本输入生成上下文相关回复。然而我们发现，它们常错误地将缺乏视觉证据的文本视为图像的一部分，导致错误回应。基于此，我们探究LVLM是否具备判断文本概念是否在图像中具象化的能力，发现一类称为视觉缺位感知（VA）的前馈网络（FFN）神经元，其具有独特的激活模式，能一致地指示视觉缺失。利用这些模式，我们构建了一个检测模块，系统性分类输入标记是否具有视觉依据。根据检测结果，我们提出一种优化方法：重解释问题提示或在生成时替换被识别为缺失的标记。大量实验表明，该方法有效缓解了模型对文本视觉存在的错误假设，并在多种LVLMs上展现出通用性。
原文摘要 · Abstract (English)
Large Vision-Language Models (LVLMs) generate contextually relevant responses by jointly interpreting visual and textual inputs. However, our finding reveals they often mistakenly perceive text inputs lacking visual evidence as being part of the image, leading to erroneous responses. In light of this finding, we probe whether LVLMs possess an internal capability to determine if textual concepts are grounded in the image, and discover a specific subset of Feed-Forward Network (FFN) neurons, termed Visual Absence-aware (VA) neurons, that consistently signal the visual absence through a distinctive activation pattern. Leveraging these patterns, we develop a detection module that systematically classifies whether an input token is visually grounded. Guided by its prediction, we propose a method to refine the outputs by reinterpreting question prompts or replacing the detected absent tokens during generation. Extensive experiments show that our method effectively mitigates the models' tendency to falsely presume the visual presence of text input and its generality across various LVLMs.
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。