Multi-hop QA Retrieval
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
81.46R@5
42
Jun 16, 2026
54.8R@2
36
Mar 17, 2026
97.2R@5
33
May 22, 2026
90.1Recall@2
32
Jun 16, 2026
47.7R@2
21
May 13, 2026
25.6PR@5
8
Mar 27, 2026
45.7PR@5
8
Mar 27, 2026
78.8Precision@5 (%)
8
Mar 27, 2026
0.791NDCG
5
Apr 1, 2026
0.834NDCG
5
Apr 1, 2026
0.908NDCG
5
Apr 1, 2026
63.6NDCG
5
Apr 1, 2026