AI RESEARCH

Bridging SFT and RL: Dynamic Policy Optimization for Robust Reasoning

arXiv CS.LG

ArXi:2604.08926v1 Announce Type: new Post-