Agentic RL 3: From Tokens to Trajectories, Long-Horizon Reinforcement Learning
What is after GRPO? Advanced long-horizon agentic reinforcement learning. (DAPO, GSPO, SAO, FlashREINFORCE, more advanced training recipes.)
What is after GRPO? Advanced long-horizon agentic reinforcement learning. (DAPO, GSPO, SAO, FlashREINFORCE, more advanced training recipes.)