S$^2$R: Teaching LLMs to Self-verify and Self-correct via Reinforcement Learning

LRM
Re-assign community