Logical reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
91.4Calibrated Accuracy
17
May 13, 2026
98.04Accuracy
17
Feb 26, 2026
4.5Pass@1
16
Jun 25, 2026
87Accuracy
16
Apr 8, 2026
94.8Recolr
16
Feb 26, 2026
94.4Recolr
16
Feb 26, 2026
68.07Accuracy
16
Feb 26, 2026
85Accuracy
16
Feb 26, 2026
70.11Accuracy
15
Apr 29, 2026
84.4Accuracy (ReClor val)
15
Feb 26, 2026
96.3Accuracy
15
May 1, 2026
81.62Accuracy
14
Jun 4, 2026
15.5Overall LRL Score
14
May 12, 2026
48.61Pass@1 Accuracy
14
Feb 26, 2026
78.2Avg@16
14
Feb 26, 2026
59.4Avg@16
14
Feb 26, 2026
69Accuracy
14
Feb 26, 2026
62.63Performance (%)
14
Feb 26, 2026
0.548Performance
14
Feb 26, 2026
95.53GSM8k Score
13
Jul 3, 2026
79.5H-Comb Accuracy
13
May 11, 2026
84Original Accuracy
13
May 11, 2026
72.9Accuracy
13
Feb 26, 2026
90.05Accuracy
12
Jun 9, 2026
59.2Performance
12
May 27, 2026