Surprisal from Larger Transformer-based Language Models Predicts fMRI Data More Poorly.

Yi-Chien Lin William Schuler Surprisal from Larger Transformer-based Language Models Predicts fMRI Data More Poorly. EACL (Volume 2: Short Papers) 2026 179-186 conf/eacl/2026-2 https://doi.org/10.18653/v1/2026.eacl-short.11 https://aclanthology.org/2026.eacl-short.11/ db/conf/eacl/eacl2026-2.html#LinS26 streams/conf/eacl