Beyond Training Objectives: Interpreting Reward Model Divergence in Large Language Models Paper • 2310.08164 • Published Oct 12, 2023 • 4
