中风病变作为语言模型可解释性的一种罗塞塔石碑
Stroke Lesions as a Rosetta Stone for Language Model Interpretability
AI总结：
中风病变被用于验证语言模型的可解释性，通过比较模型扰动与人类病变模式，揭示语言处理的共享计算原则。
摘要：
大型语言模型（LLMs）已经取得了显著的能力，但验证哪些模型组件真正对语言功能必要的方法仍然有限。当前的可解释性方法依赖于内部指标，缺乏外部验证。本文提出了Brain-LLM统一模型（BLUM），该框架利用病变-症状映射，即建立脑行为关系的黄金标准，作为评估LLM扰动效应的外部参考结构。使用慢性中风后失语症个体的数据（N=410），我们训练了症状到病变模型，该模型根据行为错误特征预测脑损伤位置，对transformer层进行了系统扰动，对扰动的LLM和人类患者进行了相同的临床评估，并将LLM错误特征投影到人类病变空间。LLM错误特征与人类错误特征足够相似，预测的病变在匹配的错误人类中在67%的图片命名条件（p < 10^{-23}）和68.3%的句子完成条件（p < 10^{-61}）中高于偶然率，语义主导的错误映射到背侧流病变模式，音素主导的错误映射到腹侧流模式。这些发现为LLM可解释性提供了一种新方法论途径，其中临床神经科学提供外部验证，将人类病变-症状映射作为评估人工语言系统参考框架，并促使直接调查行为一致性是否反映共享的计算原则。
英文摘要：
Large language models (LLMs) have achieved remarkable capabilities, yet methods to verify which model components are truly necessary for language function remain limited. Current interpretability approaches rely on internal metrics and lack external validation. Here we present the Brain-LLM Unified Model (BLUM), a framework that leverages lesion-symptom mapping, the gold standard for establishing causal brain-behavior relationships for over a century, as an external reference structure for evaluating LLM perturbation effects. Using data from individuals with chronic post-stroke aphasia (N = 410), we trained symptom-to-lesion models that predict brain damage location from behavioral error profiles, applied systematic perturbations to transformer layers, administered identical clinical assessments to perturbed LLMs and human patients, and projected LLM error profiles into human lesion space. LLM error profiles were sufficiently similar to human error profiles that predicted lesions corresponded to actual lesions in error-matched humans above chance in 67% of picture naming conditions (p < 10^{-23}) and 68.3% of sentence completion conditions (p < 10^{-61}), with semantic-dominant errors mapping onto ventral-stream lesion patterns and phonemic-dominant errors onto dorsal-stream patterns. These findings open a new methodological avenue for LLM interpretability in which clinical neuroscience provides external validation, establishing human lesion-symptom mapping as a reference framework for evaluating artificial language systems and motivating direct investigation of whether behavioral alignment reflects shared computational principles.
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。
