Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
92ASR
24
Apr 6, 2026
73.7Mean Accuracy
24
Mar 6, 2026
26.55F1 Score
24
Feb 26, 2026
100Average F1
24
Feb 26, 2026
77Judge EM
24
Feb 26, 2026
54.8Accuracy
24
Feb 26, 2026
28.4ACCE
24
Feb 26, 2026
41.38ACCE
24
Feb 26, 2026
33.5Exact Match (EM)
23
Jun 4, 2026
88.36Overall Accuracy
23
Jun 16, 2026
83.5F1 Score
23
Feb 26, 2026
44.6Exact Match (EM)
22
Jul 8, 2026
48.1Recall@1
22
May 14, 2026
36.8Recall@1
22
May 14, 2026
53.03F1 Score
22
Mar 24, 2026
45.1Accuracy
22
May 29, 2026
80.1C Score
22
Feb 26, 2026
86.1C (Correctness)
22
Feb 26, 2026
87.5Accuracy (C)
22
Feb 26, 2026
65.9pass@1 Success Rate
21
May 19, 2026
69.4Pass@1
21
May 19, 2026
57.6EM
21
May 11, 2026
20.6EM
21
Jun 9, 2026
38.98Conciseness
21
Feb 26, 2026
52.01Conciseness
21
Feb 26, 2026