Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
45Exact Match
2
Feb 26, 2026
77.93Top-1 Accuracy
2
Feb 26, 2026
36.6Accuracy
2
Feb 26, 2026
51.9F1 Score
2
Feb 26, 2026
1Task 1: Single Supporting Fact
2
Feb 26, 2026
63.21Accuracy
1
Jul 1, 2026
77.48Accuracy
1
Jul 1, 2026
26.2Text EM
1
Jun 30, 2026
53.53Answer Accuracy
1
Jun 26, 2026
2Selected Count
1
Jun 23, 2026
26Selection Rate
1
Jun 23, 2026
6Selection Rate
1
Jun 23, 2026
0Selection Rate
1
Jun 23, 2026
72Winrate
1
Jun 23, 2026
66Winrate
1
Jun 23, 2026
1.7Delta TQA Performance (pp)
1
Jun 15, 2026
22.6F1 Score
1
Jun 11, 2026
0Accuracy
1
Jun 4, 2026
0Calibration Rate
1
Jun 4, 2026
54Correctness Rate (CR)
1
Jun 4, 2026
0Accuracy (CR)
1
Jun 4, 2026
49Correctness Rate (CR)
1
Jun 4, 2026
79.8Accuracy
1
May 29, 2026
27Accuracy
1
Apr 8, 2026
17.5Accuracy
1
Apr 8, 2026