arXiv cs.AI· Camila Kolling, Omer Moussa, Emin \c{C}elik, Mariya Toneva·· 2 天前AI 评分29
SCRIBE：可扩展的可解释脑语义上下文表征框架
SCRIBE: Scalable Contextual Representations of Interpretable Brain-Grounded Semantics
AI 导读
研究者提出 SCRIBE，一个基于神经科学语义空间、用开源大语言模型集成生成可解释词级语义表征的框架。在 921 名参与者的行为研究中，SCRIBE 捕捉到人类语义特征判断随语境的变化，达到估计人类噪声上限的 75-81%；在自然故事聆听的脑活动数据上，其预测语言选择性脑区神经响应的表现显著优于非上下文语义表征。
正文 · 原文
Abstract:Understanding how the human brain constructs meaning from real-world language remains a fundamental challenge, as semantics depend on context that varies in natural settings. To enable the study of real-world linguistic meaning in the brain, we introduce SCRIBE, a framework grounded in a neuroscience-informed semantic space that leverages an ensemble of open-source large language models to generate interpretable, word-level semantic representations of arbitrary text. In a behavioral study with 921 participants, we show that human semantic feature judgments vary systematically with context, and that SCRIBE captures these shifts, reaching 75-81% of the estimated human noise ceiling. Using brain activity recorded while participants listen to naturalistic stories, we further demonstrate that SCRIBE predicts neural responses in language-selective brain regions, significantly outperforming non-contextual semantic representations. Variance partitioning reveals that SCRIBE explains unique brain-predictive information beyond syntactic features, suggesting that its representations capture aspects of linguistic meaning not accounted for by the measured syntactic features. Together, these findings establish SCRIBE as a scalable and interpretable foundation for studying natural language semantics in the brain.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.21518 [cs.AI] |
| (or arXiv:2610.21518v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.21518 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Camila Kolling [view email]
[v1]
Thu, 15 Oct 2026 08:51:15 UTC (8,125 KB)
来源：arXiv cs.AI · arxiv.org