Single-hop QA
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
72EM
52
May 13, 2026
48.5Accuracy
17
Mar 20, 2026
43EM
13
Feb 26, 2026
42Token-level F1
9
May 20, 2026
0.87Accuracy
9
Feb 26, 2026
60.7Accuracy
5
Jun 5, 2026
44Accuracy
5
Jun 5, 2026
82.1Accuracy
5
Mar 4, 2026