Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs Paper • 2509.25779 • Published Sep 30 • 16
