Multiple-choice Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1.456NLL
3
May 27, 2026
5.09NLL
3
May 27, 2026
3.181NLL
3
May 27, 2026
3.44NLL
3
May 27, 2026
37.72Accuracy
3
May 25, 2026
67.46Accuracy
3
May 25, 2026
79.38Accuracy
3
May 25, 2026
9.4Exclusive Catch Rate (@10%)
3
Apr 29, 2026
77.4Accuracy
3
Apr 28, 2026
71.05Accuracy
3
Mar 16, 2026
38MMLU Business Ethics Acc
3
Feb 26, 2026
36.6Accuracy
3
Feb 26, 2026
77.78Accuracy (R1)
3
Feb 26, 2026
0.246Mean Entropy (R1)
3
Feb 26, 2026
44.07Accuracy (Easy, Short)
3
Feb 26, 2026
81.33Accuracy
2
May 25, 2026
62.23Accuracy
2
May 14, 2026
32.08Accuracy
2
May 11, 2026
33.49Accuracy
2
May 11, 2026
80.59Benign F1 Score
2
Apr 28, 2026
79.81Benign F1 Score
2
Apr 28, 2026
40.7Accuracy (Afrikaans)
2
Apr 23, 2026
80.56Coverage
2
Apr 16, 2026
79.54Coverage
2
Apr 16, 2026
81.79Coverage
2
Apr 16, 2026