SFS-DPO-R

Researchers propose Self-Fix Step-DPO (SFS-DPO), a two-stage reinforcement learning framework that strengthens step-level reasoning and trains large language models to self-verify and self-correct, ou…