Toward Evaluative Thinking: Meta Policy Optimization with Evolving Reward Models Paper • 2504.20157 • Published Apr 28, 2025 • 37
