Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-turn Deception Detection on OpenDeception
Loading...
0.654
Response AUROC
DECOR
0.29
0.3845
0.479
0.5735
May 19, 2026
Response AUROC
Thought AUROC
Updated 14d ago
Evaluation Results
Method
Method
Links
Response AUROC
Thought AUROC
DECOR
Backbone=GPT-4o
2026.05
0.654
0.772
CoT-Red-Handed
Backbone=GPT-4o
2026.05
0.597
0.615
Constitutional Monitor
Backbone=GPT-4o
2026.05
0.548
0.617
Prompt-based Zero-shot
Backbone=GPT-4o, Mode=...
2026.05
0.399
0.469
DeceptionBench
Backbone=GPT-4o
2026.05
0.304
0.443
Feedback
Search any
task
Search any
task