Finding Where the Buck Stops: An Automated Failure Attribution-Based Reflection Framework for Multi-Agent Collaboration
尚未评估已列入计划指标结果claim-profa-attribution

ProFA achieves the strongest reported agent-level and step-level failure-attribution accuracies among the listed methods on the held-in HotPotQA, ChartQAPro, and Mind2Web datasets and on the held-out Algorithm-Generated and Hand-Crafted subsets.

来源:paper-fixed:Section 4.5; Table 1; PDF page 7

报告指标与观测值

agent_level_accuracy

reported-profa-table-1

论文报告 0.85 fraction

实际观测 — fraction

step_level_accuracy

reported-profa-table-2

论文报告 0.53 fraction

实际观测 — fraction

step_level_accuracy

reported-profa-table-3

论文报告 0.2 fraction

实际观测 — fraction

Assessment(0)

这条 Claim 暂无已发布的不可变 Assessment。

关联运行(0)

这条 Claim 暂无关联执行。