Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
73.8Accuracy
19
Mar 24, 2026
46.4Exact Match
18
Jun 4, 2026
22.3Exact Match
18
Jun 4, 2026
69.6pass@1
18
May 20, 2026
32pass@1
18
May 20, 2026
58pass@1
18
May 20, 2026
46.9Exact Match (EM)
18
May 19, 2026
83.2F1 Score
18
May 19, 2026
61.13EM
18
Feb 26, 2026
42.4EM
18
Feb 26, 2026
16.1EM
18
Feb 26, 2026
38.2EM
18
Feb 26, 2026
47.1Exact Match (EM)
18
Jul 8, 2026
71F1 Score (Bamb)
17
Jun 12, 2026
61.3F1 Score
17
Jun 12, 2026
79.1Accuracy (2WikiMultiHopQA)
17
Jun 11, 2026
31Success Rate (SR)
17
May 18, 2026
72SR
17
May 18, 2026
78SR
17
May 18, 2026
52Success Rate (SR)
17
May 18, 2026
34.3SubEM
17
Apr 21, 2026
73.4SubEM
17
Apr 21, 2026
65.7SubEM
17
Apr 21, 2026
68.2F1 Score
17
Apr 10, 2026
49.2F1
17
Apr 10, 2026