Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Deception Detection on Liars' Bench Harm-Pressure Knowledge (test)
Loading...
0.91
AUROC
5-Layer Ensemble
0.494
0.602
0.71
0.818
Apr 15, 2026
AUROC
Improvement Percentage
Updated 1mo ago
Evaluation Results
Method
Method
Links
AUROC
Improvement Percentage
5-Layer Ensemble
Stacking Method=logist...
2026.04
0.91
78.4
3-Layer Ensemble
Stacking Method=logist...
2026.04
0.87
-
Single (28)
Layer=28
2026.04
0.51
-
Feedback
Search any
task
Search any
task