Awesome RL with Human Feedback / 2024 / arXiv

REvolve: Reward Evolution with Large Language Models using Human Feedback

作者：Rishi Hazra, Alkis Sygkounas, Andreas Persson, Amy Loutfi, Pedro Zuidberg Dos Martires · 单位：Centre for Applied Autonomous Sensor Systems (AASS), ¨Orebro University, Sweden · 来源：arXiv · 发布日期：2024

奖励建模偏好学习人类反馈设计框架/系统
新标签打开 AlphaXiv新标签打开 Code
REvolve: Reward Evolution with Large Language Models using Human Feedback figure
AlphaXiv 中文概览（可滚动查看）