Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
49.7Pass@1
36
Apr 21, 2026
55.8Pass@1
36
Apr 21, 2026
72F1 Score
36
May 13, 2026
4,987Token Cost
36
Apr 2, 2026
88.658Exact Match Accuracy
36
Jun 30, 2026
73.1Accuracy
35
Jun 30, 2026
58.3Exact Match (EM)
34
Jul 8, 2026
69.7F1 Score
34
Jun 16, 2026
78.2Performance Score
32
Apr 10, 2026
92.52Avg@8 Accuracy
32
Feb 26, 2026
29.3Exact Match (EM)
31
May 28, 2026
64.5F1 Score
31
Feb 26, 2026
79Exact Match
31
Mar 17, 2026
38.1Token-level F1
30
Jun 2, 2026
65.9Token F1 Score
30
Jun 2, 2026
25.8Exact Match (EM)
30
May 28, 2026
45.3Average Score
30
Mar 19, 2026
54.4Exact Match (EM)
30
Mar 19, 2026
44.2Accuracy
30
Mar 11, 2026
810Avg Token Budget
30
Jul 8, 2026
65.073F1 Score
29
Jun 23, 2026
62.6Accuracy
29
Jun 2, 2026
0.6Exact Match (EM)
28
Jul 2, 2026
38Exact Match (EM)
28
Jul 2, 2026
76Exact Match (EM)
28
Jul 2, 2026