Multihop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
60EM
36
Jul 2, 2026
42EM
36
Jul 2, 2026
75EM
36
Jul 2, 2026
70.3Accuracy (8K Context)
6
May 12, 2026
59.4Accuracy (8K Context)
6
May 12, 2026
62.5F1
6
Mar 4, 2026
0.901Performance
5
Feb 26, 2026