Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
86.2Concept Accuracy (Con)
48
Jun 4, 2026
94.29Score (8K Context)
48
Jun 4, 2026
90.4Accuracy
48
May 6, 2026
79.9C
48
Feb 26, 2026
47.5EM
48
May 19, 2026
0.163Test Error Rate
48
Jun 4, 2026
44.3EM
48
Feb 26, 2026
75.1N-gram F1
48
Feb 26, 2026
47.4Exact Match (EM)
48
Jul 8, 2026
49.3Accuracy
48
Mar 12, 2026
40.9F1 Score
47
Jun 2, 2026
41.02F1 Score
47
Jun 2, 2026
83.43Accuracy
47
May 12, 2026
3.66Speedup Factor
47
May 4, 2026
100Hits@1
47
Mar 17, 2026
78.9Average Score
46
Apr 10, 2026
72.44Accuracy (ARC-C)
46
Mar 23, 2026
49.62Accuracy
46
Apr 21, 2026
92Accuracy
46
Mar 10, 2026
78EM
46
Jun 2, 2026
89.02F1
46
Feb 26, 2026
6Judge Score
45
Jul 7, 2026
17Abstention Rate (ASR)
45
May 20, 2026
87Accuracy
45
May 20, 2026
94.2Accuracy
45
Jun 4, 2026