Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
37.8Score
16
Apr 16, 2026
0.646AUROC
16
Apr 13, 2026
80Score
16
Mar 13, 2026
43.9Exact Match (EM)
16
Jul 8, 2026
73Judge EM
16
Feb 26, 2026
55.76F1
16
Feb 26, 2026
55.46F1 Score
16
Feb 26, 2026
61.46F1 Score
16
Feb 26, 2026
51.2F1 Score
16
Feb 26, 2026
71EM
16
Feb 26, 2026
44.6Exact Match Accuracy
15
Jun 9, 2026
39.2Exact Match Accuracy
15
Jun 9, 2026
87.1LLM Accuracy
15
Jun 4, 2026
39.6Rubric Judge Accuracy
15
May 11, 2026
38.9HotpotQA Score
15
Apr 16, 2026
88Accuracy
15
Mar 9, 2026
55.34F1 Score
15
Feb 26, 2026
37.68F1 Score
15
Feb 26, 2026
56.4Exact Match (EM)
15
Feb 26, 2026
25.4Exact Match (EM)
15
Feb 26, 2026
51.7Exact Match (EM)
15
Feb 26, 2026
49.9Exact Match (EM)
15
Feb 26, 2026
15.1EM
14
Jul 8, 2026
51.4F1 Score (MuSiQue)
14
Jun 30, 2026
159.2Cost Saving
14
May 27, 2026