Step-level reasoning verification
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
44PR-AUC
19
Apr 23, 2026
52.7PR-AUC
19
Apr 23, 2026
69.8PR-AUC
19
Apr 23, 2026
73.4PR-AUC
19
Apr 23, 2026
82.5PR-AUC
19
Apr 23, 2026
68.2PR-AUC
19
Apr 23, 2026
93PR-AUC
19
Apr 23, 2026
37.3ID Score
13
Apr 23, 2026