😼Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback

https://arxiv.org/abs/2307.15217

中文解读：https://mp.weixin.qq.com/s/BCdX6PuEdSR7D3WJ4ffonA