Logical reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
81.4Accuracy
9
Feb 26, 2026
93.19Accuracy
8
Jun 9, 2026
60.22Accuracy
8
May 29, 2026
32Accuracy (*)
8
Apr 14, 2026
50.67Accuracy
8
Apr 14, 2026
39.33Accuracy (*)
8
Apr 14, 2026
0.686Accuracy
8
Apr 8, 2026
79.9Accuracy
8
Apr 8, 2026
71.95Accuracy (ProofWriter d5 OWA)
8
Apr 8, 2026
73.5Accuracy
8
Apr 2, 2026
90Accuracy
8
Apr 2, 2026
61.35EM
8
Mar 24, 2026
79.09Accuracy
8
Feb 26, 2026
68.4Accuracy
8
Feb 26, 2026
78.5Accuracy
7
Jun 2, 2026
42.57Accuracy
7
May 21, 2026
25.2Accuracy
7
May 21, 2026
86.51Accuracy
7
Apr 21, 2026
99.4Accuracy
7
Apr 21, 2026
81.3Accuracy
7
Mar 20, 2026
56.4Hit@1 (LOGIQA)
7
Feb 26, 2026
86.2Exact Match Accuracy
7
Feb 26, 2026
54.9Score (Test-All)
7
Feb 26, 2026
57.2Accuracy (Test-All)
7
Feb 26, 2026
0.566Accuracy (Test-All)
7
Feb 26, 2026