arXiv cs.CL· Yue Huang, Yanbo Wang, Lichi Li, Yuanyuan Chen, Andreas Haupt, Xiaonan Luo, Zheyuan Liu, Jihye Choi, Felix Juefei-Xu, Xiangliang Zhang, Radha Poovendran, Misha Sra, Ying Liu, Ben Burtenshaw, Thomas Wolf, Alex Pentland, Zichen Chen, Zhangchen Xu·· 4 小时前AI 评分57
RSI Arena：借助人类反馈迈向递归自我改进的自主后训练研究
RSI Arena: Towards Recursive Self-Improvement with Human Feedback
AI 导读
RSI Arena 让十个前沿模型担任训练智能体，在同一 64-GPU 集群上自主完成后训练，全部产出了可用的助手模型。
来源：arXiv cs.CL · arxiv.org