Failure Detection and Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
73.3Detection Accuracy
8
Feb 26, 2026
91.7Detection Accuracy
8
Feb 26, 2026
65Detect Acc.
6
Feb 26, 2026
78.8Detection Accuracy
6
Feb 26, 2026
72.5Detection Accuracy
2
Feb 26, 2026
99Detection Accuracy
2
Feb 26, 2026