Awesome RL with Human Feedback / 2024 / arXiv

On Diversified Preferences of Large Language Model Alignment

作者：Dun Zeng, Yong Dai, Pengyu Cheng, Tianhao Hu, Wanshun Chen, Nan Du, Zenglin Xu · 单位：Tencent AI Lab, Alibaba Group, Peng Cheng Lab · 来源：arXiv · 发布日期：2024

奖励建模偏好学习对齐目标设计多目标对齐评测/基准
新标签打开 AlphaXiv新标签打开 Code
On Diversified Preferences of Large Language Model Alignment figure
AlphaXiv 中文概览（可滚动查看）