Step-level correctness assessment
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
55.3PR-AUC
22
Apr 23, 2026
39.5PR-AUC
22
Apr 23, 2026
0.789PR-AUC
22
Apr 23, 2026
81.9PR-AUC
22
Apr 23, 2026
79.9PR-AUC
22
Apr 23, 2026
32.9PR-AUC
22
Apr 23, 2026
70.2PR-AUC
22
Apr 23, 2026
58.6PR-AUC
22
Apr 23, 2026