Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization

Re-assign community