LLM语言网络：一种识别因果任务相关单元的神经科学方法
The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units
AI总结：
本文借鉴神经科学方法，在18个流行LLMs中识别出语言选择性单元，证明其对语言任务的因果作用及与人类语言系统大脑记录的贴合性，并探讨该方法在其他认知领域的适用性。
摘要：
大型语言模型（LLMs）不仅在语言任务上表现出卓越能力，还在逻辑推理和社会推理等非语言性质的各类任务上表现出色。在人类大脑中，神经科学已识别出一个选择性且因果性支持语言处理的核心语言系统。本文探讨LLMs中是否出现类似的语言特化现象。我们采用神经科学中相同的定位方法，在18个流行LLMs中识别出语言选择性单元。随后，通过证明消融LLM语言选择性单元（而非随机单元）会导致语言任务性能急剧下降，确立了这些单元的因果作用。相应地，语言选择性LLM单元比随机单元更贴合人类语言系统的大脑记录。最后，我们研究该定位方法是否适用于其他认知领域：尽管在部分LLMs中发现了推理和社会能力的特化网络，但模型间存在显著差异。这些发现为大型语言模型的特化提供了功能和因果证据，并凸显了与大脑功能组织的相似性。
英文摘要：
Large language models (LLMs) exhibit remarkable capabilities on not just language tasks, but also various tasks that are not linguistic in nature, such as logical reasoning and social inference. In the human brain, neuroscience has identified a core language system that selectively and causally supports language processing. We here ask whether similar specialization for language emerges in LLMs. We identify language-selective units within 18 popular LLMs, using the same localization approach that is used in neuroscience. We then establish the causal role of these units by demonstrating that ablating LLM language-selective units -- but not random units -- leads to drastic deficits in language tasks. Correspondingly, language-selective LLM units are more aligned to brain recordings from the human language system than random units. Finally, we investigate whether our localization method extends to other cognitive domains: while we find specialized networks in some LLMs for reasoning and social capabilities, there are substantial differences among models. These findings provide functional and causal evidence for specialization in large language models, and highlight parallels with the functional organization in the brain.
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。
