AI RESEARCH

Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO

arXiv CS.AI

ArXi:2505.11595v5 Announce Type: replace-cross Reinforcement learning (RL) has proven effective in strengthening the reasoning capabilities of large language models (LLMs). A widely adopted method, Group Relative Policy Optimization (GRPO), has shown strong empirical results in