Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
41.3EM
3
May 14, 2026
43.1EM
3
May 14, 2026
45.4Accuracy
3
Apr 21, 2026
31.3Exact Match (EM)
3
Mar 31, 2026
92Avg F1
3
Feb 26, 2026
29.91Accuracy
3
Feb 26, 2026
72.88Answer EM
3
Feb 26, 2026
67.08Accuracy
3
Feb 26, 2026
72.1Accuracy
3
Feb 26, 2026
34.2Accuracy
3
Feb 26, 2026
0.621Planner
2
Jul 2, 2026
46.96Planner Score
2
Jul 2, 2026
14.5EM (%)
2
Jun 16, 2026
7.67F1 Difference (pp)
2
Jun 16, 2026
61Pass@1
2
May 19, 2026
61.4Pass@1
2
May 19, 2026
35.6GraphRAG F1
2
Apr 21, 2026
1PTR Accuracy
2
Apr 7, 2026
1Avg F1
2
Feb 26, 2026
20.98Accuracy
2
Feb 26, 2026
0.306Accuracy
2
Feb 26, 2026
45.8Exact Match (EM)
2
Feb 26, 2026
22.1Exact Match (EM)
2
Feb 26, 2026
74.27Accuracy
1
Apr 9, 2026
92.6Accuracy
1
Apr 7, 2026