Long-context Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
55.32F1 (Multi Hop)
174
Jun 23, 2026
90.3SH Score
84
May 21, 2026
85.2Accuracy
72
Feb 26, 2026
7,011HotpotQA
69
Mar 24, 2026
59.71HotPotQA Accuracy
59
Jun 30, 2026
63.9Accuracy
54
Feb 26, 2026
31.5F1 Score
45
May 19, 2026
97.1Single-Hop LLJ Score
45
May 12, 2026
75.5Accuracy
38
May 8, 2026
80Accuracy
38
May 8, 2026
53.56F1 Score
38
Feb 27, 2026
36.75SubEM
36
Apr 10, 2026
22SubEM
36
Apr 10, 2026
29.33SubEM
36
Apr 10, 2026
79.5SubEM
36
Apr 10, 2026
56.77Overall Accuracy
33
May 8, 2026
73.54Avg@4 Score
22
Feb 26, 2026
65.49Mean Score
21
Feb 26, 2026
51.02F1 Score
19
Apr 10, 2026
52Accuracy
18
Jul 3, 2026
25Accuracy
18
Jul 3, 2026
21Accuracy
18
Jul 3, 2026
23Accuracy
18
Jul 3, 2026
25Accuracy
18
Jul 3, 2026
70Accuracy
18
Jul 3, 2026