Awesome RL with Human Feedback / 2023 / arXiv

Direct Preference-based Policy Optimization without Reward Modeling

作者：Gaon An, Junhyeok Lee, Xingdong Zuo, Norio Kosaka, Kyung-Min Kim, Hyun Oh Song · 单位：Seoul National University · 来源：arXiv · 发布日期：2023

奖励建模偏好学习策略优化无奖励对齐离线人类反馈强化学习
新标签打开 AlphaXiv
Direct Preference-based Policy Optimization without Reward Modeling figure
AlphaXiv 中文概览（可滚动查看）