arXiv cs.AI· Zhiyong Cao, Hengjie Zhu, Yongxin Wang, Xuchen Li, Jintao Li, Shuxia Liu, Dunqiang Liu, Bo Wang, Kairan Zhang, Cong Lin, JiangDi Yu, Siyuan Liu, Sijia Yao, Yi-Liang Wei, Lingxu Ma, Aijia Wang, Anxin Chen, Zixian Wei, Xiang Ruan, Ruizhi Xu, Ruqian Shi, Baihua Xiao, Sicheng Zhou·· 8 天前AI 评分43
Trial by Trace：AI Agent 运行框架可靠性与故障归因评测基准
Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution
AI 导读
研究人员推出 TraceTrial-Bench 评测基准，通过 95 个专家级任务评估 AI Agent 运行框架（harness）的可靠性与故障归因。该基准涵盖上下文、执行、治理等 7 个可靠性维度，并设计了轨迹追踪（TT）与故障归因（FA）模块。
来源：arXiv cs.AI · arxiv.org