Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
64.2CEM
12
Apr 14, 2026
5,583Token Count
12
Apr 2, 2026
17,910Output Tokens
12
Apr 2, 2026
48.15F1 Score
12
Mar 20, 2026
45.6Accuracy
12
Mar 11, 2026
53Accuracy
12
Mar 11, 2026
22.6Accuracy
12
Mar 11, 2026
98Avg F1
12
Feb 26, 2026
36.3EM
12
Feb 26, 2026
90.08Support F1
12
May 8, 2026
1.86Helmet Score
11
Apr 21, 2026
1Average F1 Score
11
Feb 26, 2026
100Recall
11
Feb 26, 2026
76.27Accuracy
10
Jul 2, 2026
49.07Accuracy
10
Jul 2, 2026
26.9Exact Match (EM)
10
Jun 5, 2026
45.3EM
10
Jun 5, 2026
51.6EM
10
Jun 2, 2026
88.4Expected Utility
10
May 14, 2026
23.62EM
10
Apr 21, 2026
47.91EM
10
Apr 21, 2026
30.81Exact Match (EM)
10
Apr 21, 2026
61EM
10
Apr 14, 2026
22.6Accuracy
10
Feb 27, 2026
31.6MHQ Accuracy
10
Feb 26, 2026