Hallucination Detection
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
88.43Accuracy
14
May 27, 2026
80.3Accuracy
14
May 27, 2026
85.34Accuracy
14
May 27, 2026
79.19Accuracy
14
May 27, 2026
76.14AUROC
14
May 27, 2026
69.9Balanced Accuracy
14
May 11, 2026
0.83AUROC (Token Entropy)
14
Apr 21, 2026
77.8AUROC (Token Entropy)
14
Apr 21, 2026
88.4AUROC
14
Apr 2, 2026
89.4AUROC
14
Apr 2, 2026
79.5Macro F1
14
Mar 4, 2026
84.5Macro F1
14
Mar 4, 2026
85.28AUC
14
May 27, 2026
77Precision
14
Feb 26, 2026
86.96HE
14
Feb 26, 2026
83.27Precision
14
Feb 26, 2026
73.89Neg. Score
14
Feb 26, 2026
98F1 Score
14
May 7, 2026
0.721AUCs
14
Feb 26, 2026
0.852AUROC
12
Jun 19, 2026
91.2AUROC
12
Jun 19, 2026
0.696AUROC
12
Jun 19, 2026
85.7AUROC
12
Jun 2, 2026
61AUROC
12
Jun 2, 2026
69.9AUROC
12
Jun 2, 2026