Awesome RL with Human Feedback / 2024 / arXiv

Training Diffusion Models with Reinforcement Learning

作者：Kevin Black, Michael Janner, Yilun Du, Ilya Kostrikov, Sergey Levine · 单位：University of California, Berkeley, Massachusetts Institute of Technology · 来源：arXiv · 发布日期：2024

文生图对齐偏好学习
新标签打开 AlphaXiv新标签打开 Code
Training Diffusion Models with Reinforcement Learning figure
AlphaXiv 中文概览（可滚动查看）