Mistake detection
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
60F-score
8
Jun 2, 2026
48.71AP
5
Mar 11, 2026
76.83AP
5
Mar 11, 2026
59.5AP
5
Mar 11, 2026
57F1
4
Jul 1, 2026
35.18F1@.5
4
Mar 27, 2026
57.55F1@.5
3
Mar 27, 2026
78.05F1@0.5
3
Mar 27, 2026