Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
62.5Accuracy
25
Feb 26, 2026
83.1Accuracy
25
Feb 26, 2026
40.1Accuracy
25
Feb 26, 2026
86.7Accuracy (CSQA)
24
Jun 30, 2026
71.4Exact Match
24
Jun 16, 2026
72.9Accuracy
24
Jun 12, 2026
94.37Accuracy
24
Jun 12, 2026
62Accuracy
24
Jun 11, 2026
38.42EM
24
Jun 11, 2026
66.75EM
24
Jun 11, 2026
36.71EM
24
Jun 11, 2026
64.92EM
24
Jun 11, 2026
44.23Accuracy
24
Jun 1, 2026
96.08Single-hop Score
24
May 28, 2026
57.7Accuracy
24
May 22, 2026
75Accuracy
24
May 12, 2026
37.623EM
24
Jun 30, 2026
74.26Precision
24
Apr 14, 2026
83.68Precision
24
Apr 14, 2026
68.4Accuracy
24
Apr 10, 2026
74.3Accuracy
24
Apr 10, 2026
47.7Score (2 Objectives)
24
Jul 1, 2026
81.91Multi-hop Accuracy (Cat1)
24
May 18, 2026
78.4FiQA Accuracy
24
Mar 31, 2026
79.62Accuracy
24
Mar 23, 2026