Open-domain Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
40.64NOB Accuracy
5
Feb 26, 2026
26.5EM
5
May 25, 2026
3.8Joint Answer-and-Evidence Score
4
May 25, 2026
7Joint Answer-Evidence Score
4
May 25, 2026
15.2Joint Answer-Evidence Score
4
May 25, 2026
26.4Joint Answer-and-Evidence Score
4
May 25, 2026
34.2Joint Answer-Evidence Score
4
May 25, 2026
24.75R@1
4
Apr 23, 2026
37.39R@1
4
Apr 23, 2026
29.23R@1
4
Apr 23, 2026
17.6EM
4
Feb 26, 2026
1BM25 Queries
4
Feb 26, 2026
3.47BM25 Queries
4
Feb 26, 2026
1BM25 Queries
4
Feb 26, 2026
37.22STR-EM
4
Feb 26, 2026
74.5Retrieval
4
Feb 26, 2026
65.1Retrieval Rate
4
Feb 26, 2026
66Accuracy
4
Feb 26, 2026
64.8Accuracy
4
Feb 26, 2026
52.2Exact Match
4
Feb 26, 2026
50.7Exact Match
4
Feb 26, 2026
20.4Accuracy
3
May 12, 2026
34.9Accuracy
3
May 12, 2026
39.2Accuracy
3
May 12, 2026
28.8Accuracy
3
May 12, 2026