General Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
69.02Exact Match
54
Apr 10, 2026
54.8Exact Match (EM)
52
Apr 16, 2026
45.2EM
51
Apr 10, 2026
58.2NQ Accuracy
40
May 8, 2026
70.9EM
33
Jun 4, 2026
46.1EM
32
Apr 21, 2026
50.9Exact Match
30
Jun 9, 2026
49.3EM
27
Jun 12, 2026
73.3EM
27
Jun 12, 2026
68.2EM
24
Feb 26, 2026
55.8EM
24
Feb 26, 2026
47.9EM
20
Jun 11, 2026
66.4Exact Match (EM)
20
Jun 11, 2026
49.2Exact Match (EM)
20
Jun 11, 2026
76.4Token-Level F1
20
Apr 21, 2026
48.8Accuracy
18
Feb 26, 2026
80.2Accuracy
18
Feb 26, 2026
64.4EM
18
Feb 26, 2026
53.4F1 Score
17
Jun 12, 2026
50.5F1 Score
17
Jun 12, 2026
78.4F1 Score
17
Jun 12, 2026
50.8F1 Score
17
Jun 12, 2026
40.4Exact Match
17
Jun 12, 2026
64.6Score
16
Apr 16, 2026
45.7Exact Match Accuracy
15
Jun 9, 2026