Awesome RL with Human Feedback / 2023 / arXiv

RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment

作者：Hanze Dong, Wei Xiong, Deepanshu Goyal, Rui Pan, Shizhe Diao, Jipeng Zhang, Kashun Shum, Tong Zhang · 单位：The Hong Kong University of Science and Technology, University of Illinois Urbana-Champaign · 来源：arXiv · 发布日期：2023

奖励建模偏好学习对齐目标设计策略优化策略梯度人类反馈强化学习
新标签打开 AlphaXiv新标签打开 Code
RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment figure
AlphaXiv 中文概览（可滚动查看）