Multi-hop Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
25.2EM
10
Feb 26, 2026
50.5EM
10
Feb 26, 2026
100Recall
10
Feb 26, 2026
30.4EM
10
Feb 26, 2026
79.8Accuracy
10
Feb 26, 2026
66LLM Accuracy
9
Jun 1, 2026
85.7LLM Acc
9
Jun 1, 2026
53.4Accuracy
9
May 22, 2026
70.33F1 Score
9
Jun 9, 2026
44.7Exact Match (EM)
9
May 13, 2026
78.6Answer Accuracy (Ans)
9
Apr 23, 2026
15.2Exact Match (EM)
9
Apr 15, 2026
75.63Accuracy
9
Apr 7, 2026
65.78Accuracy
9
Apr 7, 2026
68.98F1 Score
9
Apr 2, 2026
69.2EM
9
Mar 5, 2026
40.62-hop F1
9
Feb 26, 2026
32.6Avg@4
9
Feb 26, 2026
67.6Avg@4
9
Feb 26, 2026
2.28Latency (s)
9
Feb 26, 2026
2.3Task latency (s)
9
Feb 26, 2026
2.29Task Latency (s)
9
Feb 26, 2026
59.2Pass@3
8
Jul 7, 2026
31.1Pass@3
8
Jul 7, 2026
64Pass@3
8
Jul 7, 2026