Exploratory Preference Optimization: Harnessing Implicit Q*-Approximation for Sample-Efficient RLHF

Open Webpage

Tengyang Xie, Dylan J. Foster, Akshay Krishnamurthy, Corby Rosset, Ahmed Hassan Awadallah, Alexander Rakhlin

Published: 01 Jan 2025, Last Modified: 01 Oct 2025ICLR 2025EveryoneRevisionsBibTeXCC BY-SA 4.0