slow-fast-policy-optimization-rl

【效率与协作】Stabilize RL for LLM reasoning via three-phase decomposition: fast inner trajectory optimization, repositioning to manage off-policy drift, slow correction for stable updates. Achieve up to 2.80-point math reasoning gains over GRPO while reducing rollouts 4.93x and wall-clock time 4.19x via improved stability without changing reward structure.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准