Common Sense Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
58.27Accuracy
27
Feb 26, 2026
78.06Normalized Accuracy
26
Jun 8, 2026
91.58Accuracy
26
Feb 26, 2026
63.430-shot Accuracy
24
Jun 4, 2026
92.29Accuracy
24
Feb 26, 2026
50Accuracy
23
Jul 3, 2026
81.6Accuracy
23
Jul 3, 2026
88Accuracy (BBH Sports)
21
Jun 2, 2026
48.91Accuracy
21
Jul 3, 2026
87.27Accuracy
21
Jul 3, 2026
90.5Accuracy
21
Jul 3, 2026
81.12PIQA
20
Feb 26, 2026
95.4Accuracy
20
Apr 1, 2026
64.6Accuracy
19
May 22, 2026
83.2Accuracy
19
Apr 1, 2026
86Accuracy
18
May 26, 2026
16Wiki PPL
18
Apr 16, 2026
84.1PIQA Accuracy
18
Feb 26, 2026
80.1BoolQ Accuracy
16
Jun 2, 2026
44.88ARC-C Accuracy
16
Feb 26, 2026
79.2Accuracy
16
Apr 1, 2026
88BoolQ Accuracy
15
Jun 2, 2026
78.09Accuracy
15
Apr 7, 2026
61Average Accuracy
15
Apr 3, 2026
72.41Average Accuracy
15
Feb 26, 2026