The Pupil Becomes the Master: Eye-Tracking Feedback for Tuning LLMs

Samuel Kiegeland, David Robert Reich, Ryan Cotterell, Lena Ann Jäger, Ethan Wilcox

Keywords: Eye-tracking, LLM, DPO

Abstract: Large language models often require alignment with explicit human preferences, which can be sparse and costly. We propose a framework to leverage eye-tracking data as an implicit feedback signal to tune LLMs for controlled sentiment generation using Direct Preference Optimization. Our study demonstrates that eye-tracking feedback can be a valuable signal for tuning LLMs. This motivates future research to investigate the impact of eye-tracking feedback on various tasks, highlighting the potential of integrating eye-tracking data with LLMs to improve their performance and alignment with human preferences.

Submission Number: 61