Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
63TruthfulQA Score
61
Mar 26, 2026
76.8Accuracy
60
May 22, 2026
68.6Accuracy
60
Feb 26, 2026
79Accuracy
60
Feb 26, 2026
96.07Accuracy
60
Jun 30, 2026
83.3Accuracy
60
Mar 4, 2026
81.1Normalized Accuracy (ARC-E)
59
May 19, 2026
61.8EM Accuracy
59
Jun 30, 2026
96.7Hit@1
59
Jul 2, 2026
96Accuracy
59
Feb 26, 2026
77.2RACC (Accuracy)
58
May 25, 2026
39.7EM
57
Jul 8, 2026
53.6EM (%)
57
Jun 30, 2026
73.45QA Zero-shot Average
57
Feb 26, 2026
88.9Accuracy
57
Feb 26, 2026
92.91Accuracy
57
May 26, 2026
44.11F1 Score
56
Jun 2, 2026
20.99Accuracy
56
Jun 18, 2026
87.8Accuracy
56
Feb 26, 2026
71.1EM
56
May 11, 2026
65.2Accuracy
55
May 19, 2026
48.2Average QA Score
55
Jul 8, 2026
84.1Normalized Accuracy
55
Jun 2, 2026
0.442EM
55
Mar 11, 2026
82.53Accuracy
54
Apr 30, 2026