Awesome RL with Human Feedback / 2026 / arXiv

Distributionally Robust Reinforcement Learning with Human Feedback

作者：Debmalya Mandal, Paulius Sasnauskas, Goran Radanovic · 单位：University of Warwick, UK, Max-Planck Institute for · 来源：arXiv · 发布日期：2026

鲁棒性/噪声反馈奖励建模偏好学习策略优化策略梯度人类反馈强化学习
新标签打开 AlphaXiv
Distributionally Robust Reinforcement Learning with Human Feedback figure
AlphaXiv 中文概览（可滚动查看）