← 研踪 PaperTrailarXiv:2611.20711 · v1
AgentFaultBench: State-Verified Evaluation of Tool-Fault Recovery
首次提交 2026-11-16 16:19
版本更新 2026-11-16 16:19
雷达观测 2026-11-17 13:07
摘要
Does recovery advice help an agent finish a task after a tool fails? AgentFaultBench evaluates LangGraph and CrewAI agents performing specified tasks in an isolated GitLab installation. We inject tool errors and check both the resulting files and issues and the required reads, verification and final response. The prompt supplies the expected issue and report text, so the task tests execution rather than content generation. In the primary comparison of 1,944 planned runs, advice changes success by -8.33, 0.00 and +1.06 percentage points relative to a neutral reminder; missing outcomes leave the latter two comparisons inconclusive. Later experiments test a revised recovery policy. GPT-4o mini improves, while Sol and Claude already complete the recoverable cases. In a separate 624-run study on six related tasks not previously used by that policy, the guided condition records 161 correct final states and one unknown outcome in 192 planned runs, versus 142/192 with neutral feedback. The gain lies between +9.90 and +10.42 pp when missing outcomes are included. Removing issue IDs gives 47/48 successes versus 46/48 with IDs; other execution information remains. A second checker agrees on 1,879 observed primary outcomes and 241 constructed test cases. No human validation was collected. The findings support testing advice against explicit recovery controls rather than assuming it helps.
作者备注:63 pages, 7 figures, including appendices. Submitted to the NAACL 2027 Industry Track
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-16 16:19 · 最新观测版本
AgentFaultBench: State-Verified Evaluation of Tool-Fault Recovery
Does recovery advice help an agent finish a task after a tool fails? AgentFaultBench evaluates LangGraph and CrewAI agents performing specified tasks in an isolated GitLab installation. We inject tool errors and check both the resulting files and issues and the required reads, verification and final response. The prompt supplies the expected issue and report text, so the task tests execution rather than content generation. In the primary comparison of 1,944 planned runs, advice changes success by -8.33, 0.00 and +1.06 percentage points relative to a neutral reminder; missing outcomes leave the latter two comparisons inconclusive. Later experiments test a revised recovery policy. GPT-4o mini improves, while Sol and Claude already complete the recoverable cases. In a separate 624-run study on six related tasks not previously used by that policy, the guided condition records 161 correct final states and one unknown outcome in 192 planned runs, versus 142/192 with neutral feedback. The gain lies between +9.90 and +10.42 pp when missing outcomes are included. Removing issue IDs gives 47/48 successes versus 46/48 with IDs; other execution information remains. A second checker agrees on 1,879 observed primary outcomes and 241 constructed test cases. No human validation was collected. The findings support testing advice against explicit recovery controls rather than assuming it helps.
查看此版本解读 →查看此版本原文 ↗