Training Language Models to Self-Correct via Reinforcement Learning

Xiaol.x - Training Language Models to Self-Correct via Reinforcement Learning