PapersSelected01:34Chinese contentLongHorizon-Harness:评测真实世界长时程智能体有人发了 LongHorizon-Harness,一个专测长时程智能体的框架,论文在 Hugging Face,想评估多步任务表现可以看。#LongHorizon-Harness#智能体#评估基准#长时程任务AK@_akhaliqSourceRead laterReadWorth followingUsefulFollow LongHorizon-Harness