用脑电波训练大模型，发现每个人有独特的激活模式。
Riding Brainwaves in LLM Space: Understanding Activation Patterns Using Individual Neural Signatures
- 用个体脑电数据训练线性探测器，映射大模型隐藏状态到个人脑活动。
- 个人专属探测器对高γ功率预测效果提升九倍，相关系数达0.183。
- 该信号稳定且不跨人迁移，适合脑机接口与个性化智能系统研究。
消费级脑电设备正进入日常用品，引发语言模型是否可适配个体神经反应的疑问。本文通过分析30名参与者阅读自然语句时的词级脑电信号（ZuCo语料库），为每人训练独立线性探测器，将冻结的Qwen 2.5 7B模型隐藏状态映射至其个体脑电功率。个体专用探测器在所有测试脑电特征上均优于单一群体探测器；高γ功率下，个体探测器相关系数rho=0.183，较群体探测器（rho=0.020，p<10^-4）提升九倍。作为负控实验，注视次数无个体优势（p=0.360），反映词长与频率而非个体认知。个体方向具有时间稳定性（分半余弦相似度0.824），跨人不可迁移（自相关0.369 vs. 异人0.143，p<10^-19），且与共享群体信号分离：去除群体成分后仍具预测能力。该个体信号集中于模型深层，随层数递增，峰值出现在28层中的第24层。结果在LLaMA 3.1 8B架构及词级混杂控制下依然成立。冻结语言模型深层包含稳定、个体特异的神经方向，为脑电驱动的个性化提供了几何基础。
原文摘要 · Abstract (English)
Consumer-grade EEG is entering everyday devices, from earbuds to headbands, raising the question of whether language models can be adapted to individual neural responses. We test this by asking whether frozen LLM representations encode person-specific EEG signals, directions in activation space that predict one person's brain activity but not another's. Using word-level EEG from 30 participants reading naturalistic sentences (ZuCo corpus), we train a separate linear probe for each person, mapping hidden states from a frozen Qwen 2.5 7B to that individual's EEG power. Person-specific probes outperform a single population probe on every EEG feature tested; for high-gamma power, the person-specific probe achieves rho = 0.183, a ninefold improvement over the population probe (rho = 0.020, p < 10^-4). A negative control, fixation count, shows no person-specific advantage (p = 0.360); fixation count reflects word length and frequency rather than individual cognition. The individual directions are temporally stable (split-half cosine = 0.824), non-transferable across people (self rho = 0.369 vs. other rho = 0.143, p < 10^-19), and distinct from the shared population signal: person-specific probes retain predictive power after the population component is removed. The person-specific signal concentrates in the model's deep layers, rising consistently with depth and peaking at Layer 24 of 28. The results are consistent across architectures (LLaMA 3.1 8B) and survive word-level confound controls. Frozen language models contain stable, person-specific neural directions in their deep layers, providing a geometric foundation for EEG-driven personalization.
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。