Jul 2026
Encoding EEG Signals to Examine Human-Like Next-Word Prediction Behaviour in Language Models
It is argued that modelling ERP patterns offers fine-grained analysis of the cognitive plausibility of various LMs during reading, and that only surprisal potentially correlates with language-processing ERPs, especially for open-class words with high semantic content.
B. M. Quach, B. Nguyen, C. Gurrin et al.
· arXiv.org · 0 citations