Wing-Yee Chow

Extracting structure from an LLM - how to improve on surprisal-based models of Human Language Processing

Prediction and reanalysis are considered two key processes that underly humans’ capacity to comprehend language in real time. Computational models capture it using Large Language Models (LLMs) and a statistical measure known as ‘surprisal’. Despite successes of LLMs, surprisal-based models face chal…