Fact Verification
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
92.1Accuracy
86
Jun 30, 2026
53.9Accuracy
72
May 20, 2026
82.1Label Accuracy
57
Feb 26, 2026
0.555True Deferral-Advice Loss
48
Mar 17, 2026
73.6Accuracy
45
May 4, 2026
79.47LA Score
32
Feb 26, 2026
93.6Exact Match (EM)
30
Jun 25, 2026
87.45Accuracy
30
Jun 2, 2026
81.63Precision
30
Feb 26, 2026
70Accuracy
28
Apr 28, 2026
84.73Accuracy
28
Feb 26, 2026
68.6F1 Score
24
Apr 8, 2026
89.07Label Accuracy
23
Feb 26, 2026
84Accuracy
21
Jun 24, 2026
95.1Accuracy
21
Jun 24, 2026
69Accuracy
21
Jun 24, 2026
79.6Accuracy
21
Jun 24, 2026
79.5Accuracy
21
Mar 27, 2026
85.02Accuracy
21
Mar 27, 2026
46.08METEOR (Q-only)
20
Jun 12, 2026
73Accuracy
20
May 1, 2026
61.1EM
18
Apr 10, 2026
53.9F1 Score
18
Mar 27, 2026
97.45Hits@1
17
Apr 15, 2026
67.44Accuracy
17
Feb 26, 2026