Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
37.3Accuracy
3
Feb 26, 2026
29.8Top-1 EM Accuracy
3
Feb 26, 2026
11.8Average Error Rate
3
Feb 26, 2026
0.363Error Rate
3
Feb 26, 2026
37.9Error Rate
3
Feb 26, 2026
75.98Exact Match
3
Feb 26, 2026
40EM
2
Jun 30, 2026
37.8Text EM
2
Jun 30, 2026
99.9Quality
2
Jun 26, 2026
0Cost (cents)
2
Jun 26, 2026
99.6Quality (%)
2
Jun 26, 2026
218Answer Count
2
Jun 23, 2026
47.8Accuracy
2
Jun 23, 2026
71.2Accuracy
2
Jun 23, 2026
72.1Accuracy
2
Jun 23, 2026
65.5Accuracy
2
Jun 23, 2026
95Accuracy
2
Jun 23, 2026
75Accuracy
2
Jun 23, 2026
0Mean Jaccard
2
Jun 19, 2026
6.8Mean Jaccard
2
Jun 19, 2026
0Mean Jaccard
2
Jun 19, 2026
68.3Content Accuracy
2
Jun 17, 2026
83Sel-AUC
2
Jun 17, 2026
21.65Accuracy/F1
2
Jun 16, 2026
74.1Accuracy/F1 Score
2
Jun 16, 2026