Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning

Open in new window