Awesome RL with Human Feedback / 2025 / arXiv

Segmenting Text and Learning Their Rewards for Improved RLHF in Language Model

作者：Yueqin Yin, Shentao Yang, Yujia Xie, Ziyi Yang, Yuting Sun, Hany Awadalla, Weizhu Chen, Mingyuan Zhou · 单位：The University of Texas at Austin · 来源：arXiv · 发布日期：2025

奖励建模策略优化策略梯度人类反馈强化学习信用分配/密集奖励框架/系统
新标签打开 AlphaXiv新标签打开 Code
Segmenting Text and Learning Their Rewards for Improved RLHF in Language Model figure
AlphaXiv 中文概览（可滚动查看）