Closed-book Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
29.9Accuracy
27
Feb 27, 2026
71.2Accuracy
24
Feb 27, 2026
43.62EM
24
Jun 4, 2026
76.12EM
24
Jun 4, 2026
18.8Accuracy
21
May 19, 2026
21.8Accuracy
14
Feb 26, 2026
40.1Accuracy
12
Feb 26, 2026
82Accuracy
12
Feb 26, 2026
54.4Accuracy
9
Feb 26, 2026
81.67LLM Score
8
Apr 13, 2026
73.2Accuracy
8
Feb 26, 2026
72.5Accuracy
7
Feb 26, 2026
21.2EM
6
Feb 26, 2026
48F1 Score
5
Feb 26, 2026
85.9EM
4
Apr 21, 2026
39.6Recall
4
Mar 18, 2026
78.4Accuracy
3
Feb 27, 2026
1.8EM
2
Feb 26, 2026
30Exact Match
1
Feb 26, 2026