Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs

Open in new window