End-to-end Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
85.5Overall Score
25
Apr 24, 2026
70.9Overall Score
25
Apr 24, 2026
47.14Accuracy
20
Feb 26, 2026
10.79EM
20
Feb 26, 2026
35.44EM
20
Feb 26, 2026
36.68EM
20
Feb 26, 2026
50.61Exact Match
12
Apr 7, 2026
74.21Exact Match
12
Apr 7, 2026
48.38Exact Match
12
Apr 7, 2026
45.09Exact Match
12
Apr 7, 2026
14EM
9
May 8, 2026
0.464EM
9
May 8, 2026
57.52Token F1 Score
9
Apr 21, 2026
56.53Token F1 Score
9
Apr 21, 2026
67.1EM
9
Feb 26, 2026
36.31EM
7
Feb 26, 2026
44.57EM
7
Feb 26, 2026
60.91EM
7
Feb 26, 2026
65.48EM
7
Feb 26, 2026
74.7Relevancy
6
Mar 26, 2026
0.317Accuracy
6
Feb 26, 2026
20.6Accuracy
6
Feb 26, 2026
0Step Regret
5
Jun 11, 2026
87.5Accuracy (Human)
4
Mar 13, 2026
52Rotation
4
Feb 26, 2026