Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
85.18Accuracy
121
Feb 26, 2026
95.2Accuracy
119
Apr 15, 2026
94.5Accuracy
117
May 11, 2026
0.71Accuracy
116
May 27, 2026
95.3Accuracy
114
May 11, 2026
87Accuracy
113
May 25, 2026
78.9F1
113
May 29, 2026
56.6EM
112
Jun 16, 2026
42.5Mean PRR
109
Jun 18, 2026
47.7EM
107
May 8, 2026
98.8Accuracy
106
Jun 16, 2026
82.9LLM-Judge Score
105
Feb 26, 2026
59Normalized Accuracy
105
Jun 2, 2026
91.2Accuracy
103
May 19, 2026
89.1Accuracy
103
Jul 8, 2026
55.6Normalized Accuracy
102
Apr 10, 2026
74.2Exact Match
101
May 22, 2026
94.34Accuracy
101
Feb 26, 2026
75.5F1 Score
101
Feb 26, 2026
0.853AUROC
98
Jun 4, 2026
94AUROC (TQA 8 additional samples)
98
Jun 4, 2026
68.4Accuracy
97
Apr 8, 2026
94.8Accuracy
96
Mar 4, 2026
81.33Accuracy
95
Feb 26, 2026
77.71Accuracy
95
Jun 26, 2026