Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
48.3EM
28
Jul 8, 2026
73Exact Match (EM)
28
Apr 10, 2026
47.6Accuracy
28
Jun 4, 2026
43.3EM
27
May 26, 2026
59.1EM
26
Jun 5, 2026
77.12Accuracy
26
Feb 26, 2026
86.7LLM Accuracy
25
Jun 12, 2026
46.1Exact Match (EM)
25
Jul 8, 2026
80.3Score
25
May 25, 2026
73.8Exact Match (EM)
25
Jul 8, 2026
56.2F1 Score
25
May 11, 2026
86.4Accuracy
25
Apr 1, 2026
61.6F1
25
Feb 26, 2026
70.1Accuracy
24
Jul 7, 2026
45.1Accuracy
24
Jul 7, 2026
59.664Accuracy
24
Apr 28, 2026
68.7Exact Match
24
Apr 21, 2026
26.5EM
24
Apr 21, 2026
63.5Exact Match
24
Apr 21, 2026
51.3Exact Match (EM)
24
Apr 21, 2026
53.4Exact Match (EM)
24
Apr 21, 2026
24.4EM
24
Apr 21, 2026
43.6Exact Match
24
Apr 21, 2026
44.7EM
24
Apr 21, 2026
47.55Accuracy
24
Apr 8, 2026