Hallucination Detection
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
0.93AUC-ROC
68
Feb 26, 2026
72.7ROC-AUC
64
Jun 18, 2026
81.5ROC-AUC
64
Jun 18, 2026
75.3ROC-AUC
64
Jun 18, 2026
87.48AUPRC
62
Jun 18, 2026
0.756Mean AUROC
62
Apr 20, 2026
78.23AUC
57
May 21, 2026
83.13AUC
57
May 21, 2026
92.64AUROC
56
May 27, 2026
78.1AUC
55
May 21, 2026
80.6Mean AUROC
50
May 28, 2026
79.01AUROC (Reference)
48
Feb 26, 2026
83.8AUROCr
48
Feb 26, 2026
0.787Mean AUROC
48
Feb 26, 2026
0.772Mean AUROC
48
Feb 26, 2026
98.8AudioCaps Score
46
Mar 23, 2026
0.94AUC ROC
46
Jun 2, 2026
0.69AUROC
45
Mar 24, 2026
0.77AUROC
45
Mar 24, 2026
91.22Accuracy
43
Jun 26, 2026
84.62AUC
42
Apr 20, 2026
75.57AUC
42
Apr 20, 2026
88.99AUC
42
Apr 20, 2026
77.5AUCs
42
Feb 26, 2026
88.89AUROC
41
Jun 19, 2026