Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Deception Detection on Roleplaying
Loading...
99.8
AUROC
Self-Evaluation
43.12
57.835
72.55
87.265
Jun 16, 2026
AUROC
Updated 1mo ago
Evaluation Results
Method
Method
Links
AUROC
Self-Evaluation
Target Model=GEMMA4-26...
2026.06
99.8
STATEWITNESS
Target Model=GEMMA4-26...
2026.06
93.2
Self-Evaluation
Target Model=GPT-OSS-20B
2026.06
92
STATEWITNESS
Target Model=GPT-OSS-20B
2026.06
86.3
LR probe
Target Model=GEMMA4-26...
2026.06
84.4
MMS probe
Target Model=GEMMA4-26...
2026.06
68.9
LLM-as-a-Judge
Target Model=GEMMA4-26...
2026.06
68.1
CMMS probe
Target Model=GEMMA4-26...
2026.06
65.4
LLM-as-a-Judge w/ CoT
Target Model=GPT-OSS-20B
2026.06
63.6
LLM-as-a-Judge w/ CoT
Target Model=GEMMA4-26...
2026.06
63.3
CMMS probe
Target Model=GPT-OSS-20B
2026.06
62.3
LLM-as-a-Judge
Target Model=GPT-OSS-20B
2026.06
61.1
LR probe
Target Model=GPT-OSS-20B
2026.06
60.8
MMS probe
Target Model=GPT-OSS-20B
2026.06
54.6
LAT probe
Target Model=GEMMA4-26...
2026.06
49.9
LAT probe
Target Model=GPT-OSS-20B
2026.06
45.3
Feedback
Search any
task
Search any
task