Skip to yearly menu bar Skip to main content


Can RLHF be More Efficient with Imperfect Reward Models? A Policy Coverage Perspective

Abstract

Chat is not available.