Finding Where the Buck Stops: An Automated Failure Attribution-Based Reflection Framework for Multi-Agent Collaboration
Not assessedPlannedMeasurementclaim-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.
Source: paper-fixed:Section 4.5; Table 1; PDF page 7
Reported and observed measurements
agent_level_accuracy
reported-profa-table-1
Reported 0.85 fraction
Observed — fraction
step_level_accuracy
reported-profa-table-2
Reported 0.53 fraction
Observed — fraction
step_level_accuracy
reported-profa-table-3
Reported 0.2 fraction
Observed — fraction
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.