Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
68.6Accuracy
28
Apr 2, 2026
85.8Accuracy
28
Mar 20, 2026
0.044ECE
28
Mar 4, 2026
44.3Unreliability Rate
28
Jun 30, 2026
81.3Exact Match AUC
28
Feb 26, 2026
88.23Accuracy
28
Mar 6, 2026
90Accuracy
28
Feb 26, 2026
39.5EM
28
Feb 26, 2026
42.5EM
28
Feb 26, 2026
52.14Exact Match
28
Feb 26, 2026
52.5F1 Score
28
Feb 26, 2026
77.8F1 Score
28
Feb 26, 2026
21.5BLEU-1
28
May 12, 2026
59.5EM
28
Feb 26, 2026
100Hits@1
28
Mar 17, 2026
68.5F1 (N-gram)
28
Feb 26, 2026
78.37ARC-e Accuracy
27
Jun 18, 2026
47.61ARC-c Score
27
Jun 18, 2026
84Accuracy
27
Jun 30, 2026
2.21Exact Match (EM)
27
May 12, 2026
5.26EM
27
May 12, 2026
2.41Exact Match (EM)
27
May 12, 2026
48.1EM (%)
27
May 8, 2026
47.82EM (%)
27
May 8, 2026
65.3EM (%)
27
May 8, 2026