Hallucination Detection
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
97.57AUPRC
4
Mar 9, 2026
85.17AUPRC
4
Mar 9, 2026
96.98AUPRC
4
Mar 9, 2026
0.03FPR
4
Feb 26, 2026
1FPR
4
Feb 26, 2026
0FPR
4
Feb 26, 2026
53.3ANLL
4
Feb 26, 2026
49.6ANLL
4
Feb 26, 2026
59.9ANLL
4
Feb 26, 2026
57.2ANLL
4
Feb 26, 2026
96.48AUC
4
Feb 26, 2026
0.94AUC
4
Feb 26, 2026
74.45AUCROC
4
Feb 26, 2026
72.9AUCROC
4
Feb 26, 2026
73.37AUCROC
4
Feb 26, 2026
78.6AUC-ROC
4
Feb 26, 2026
47.3Macro F1 Score
4
Feb 26, 2026
70.4AUC
3
Jul 7, 2026
64.1AUC
3
Jul 7, 2026
68.9IoU
3
Jul 2, 2026
72.4IoU
3
Jul 2, 2026
100Improvement
3
Jun 9, 2026
59.54Accuracy
3
May 28, 2026
73.89Unsafe F1 Score
3
May 6, 2026
99AUC
3
Feb 26, 2026