Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Hallucination monitoring on Inheritance domain
Loading...
90
AUROC
HalMit
59.84
67.67
75.5
83.33
Jul 21, 2025
AUROC
AUC-PR
F1
Accuracy
Updated 19d ago
Evaluation Results
Method
Method
Links
AUROC
AUC-PR
F1
Accuracy
HalMit
Backbone=Llama3.1
2025.07
90
86
82
88
SCG
Backbone=Llama3.1
2025.07
85
84
85
85
PP
Backbone=Llama3.1
2025.07
71
79
72
71
SCG
Backbone=Llama2
2025.07
70
73
68
75
HalMit
Backbone=Llama2
2025.07
70
79
80
78
ICL
Backbone=Llama2
2025.07
69
67
59
56
PP
Backbone=Llama2
2025.07
68
60
77
75
ICL
Backbone=Llama3.1
2025.07
61
73
65
65
Feedback
Search any
task
Search any
task