企业Agent新基准EnterpriseClawBench:真实场景性能仅0.663
EnterpriseClawBench: Benchmarking Agents from Real Workplace Sessions 作者:Jincheng Zhong, Weizhi Wang, Che Jiang, Kai Tian, Zhenzhao Yuan, Junlin Yang, Dianqiao Lei, Kaiyan Zhang 核心发表机构:Horizon Research、Frontis.AI 论文链接&…
2026/6/25 18:48:14