Commonsense Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
36.55Accuracy
3
Mar 6, 2026
64Score
3
Feb 27, 2026
3.9Average Relative Improvement
3
Feb 26, 2026
2.79Avg Relative Improvement
3
Feb 26, 2026
3.81Average Relative Improvement
3
Feb 26, 2026
0.761Pass@1
3
Feb 26, 2026
60Accuracy
3
Feb 26, 2026
92.98Accuracy
3
Feb 26, 2026
86Accuracy
3
Feb 26, 2026
77.94Accuracy
3
Feb 26, 2026
66.7Accuracy
3
Feb 26, 2026
0.668Accuracy
3
Feb 26, 2026
16.3Accuracy
3
Feb 26, 2026
94Accuracy
3
Feb 26, 2026
57Accuracy
2
Jun 23, 2026
41.1Accuracy
2
May 27, 2026
53.43Accuracy
2
May 19, 2026
71.7Score
2
May 15, 2026
49.4Average Accuracy
2
May 15, 2026
88.57Accuracy
2
Apr 23, 2026
24.2pass@1
2
Apr 21, 2026
0.67Log-likelihood Score
2
Apr 2, 2026
79.3Normalized Log Accuracy
2
Mar 18, 2026
0.012MAE
2
Feb 26, 2026
77Accuracy
2
Feb 26, 2026