Commonsense Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
79.7Accuracy
13
May 18, 2026
84.4Accuracy
12
May 12, 2026
50.97Reusability Score
12
Feb 26, 2026
70.1Accuracy
11
Feb 26, 2026
63.8Accuracy
10
Feb 26, 2026
44Accuracy
9
Apr 7, 2026
33.96Score
8
Feb 26, 2026
41.13Accuracy
8
Feb 26, 2026
55.18Accuracy
8
Feb 26, 2026
58.72Accuracy
8
Feb 26, 2026
83.7Accuracy
7
Feb 26, 2026
52.5Accuracy
6
Jun 24, 2026
71.07Accuracy
6
Apr 23, 2026
39.12Score
6
Apr 1, 2026
72.1Accuracy
6
Feb 26, 2026
77.62AUROC
5
Jun 5, 2026
70.7Accuracy
5
Jun 1, 2026
73.5Accuracy (n=15)
5
May 21, 2026
36.8mCSQA Accuracy (ZH)
5
Apr 21, 2026
59.4Accuracy
5
Feb 26, 2026
53.04Accuracy
5
Feb 26, 2026
49.3Accuracy
4
Jun 17, 2026
67.78Average Accuracy
4
Feb 26, 2026
67.3Accuracy
4
Feb 26, 2026
72.8Accuracy
4
Feb 26, 2026