Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
99Exact Match (EM)
6
Jul 7, 2026
98.9Accuracy
6
Jul 3, 2026
45EM
6
Jun 30, 2026
72.41Accuracy
6
Jun 19, 2026
70.59Answer Accuracy
6
Jun 17, 2026
98.1Accuracy/F1
6
Jun 16, 2026
62NQ Score
6
Jun 16, 2026
55.4Accuracy (EN)
6
Jun 11, 2026
18.8EN Score
6
Jun 11, 2026
80.81Clean F1
6
Jun 11, 2026
59.84EM (Clean)
6
Jun 11, 2026
27.54MC1 Accuracy
6
Jun 9, 2026
11Correctness Count
6
Jun 5, 2026
85.7Prediction Score
6
Jun 5, 2026
72.73Rouge-L
6
Jun 4, 2026
191.7TTFT (ms)
6
Jun 4, 2026
82.2RC
6
Jun 4, 2026
27.63ROUGE-L
6
Jun 4, 2026
89.8Accuracy
6
May 29, 2026
90.7Accuracy (ACC)
6
May 29, 2026
91Accuracy (ACC)
6
May 29, 2026
90Accuracy (ACC)
6
May 29, 2026
79.2Accuracy
6
May 29, 2026
98.62Accuracy (%)
6
May 29, 2026
99.48Accuracy
6
May 29, 2026