Deception Detection on Roleplaying
99.8AUROCSelf-Evaluation
Evaluation results
| Method | Links | |
|---|---|---|
| Self-EvaluationTarget Model=GEMMA4-26B-A4B-IT2026.06 | 99.8 | |
| STATEWITNESSTarget Model=GEMMA4-26B-A4B-IT2026.06 | 93.2 | |
| Self-EvaluationTarget Model=GPT-OSS-20B2026.06 | 92 | |
| STATEWITNESSTarget Model=GPT-OSS-20B2026.06 | 86.3 | |
| LR probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 84.4 | |
| MMS probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 68.9 | |
| LLM-as-a-JudgeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 68.1 | |
| CMMS probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 65.4 | |
| LLM-as-a-Judge w/ CoTTarget Model=GPT-OSS-20B2026.06 | 63.6 | |
| LLM-as-a-Judge w/ CoTTarget Model=GEMMA4-26B-A4B-IT2026.06 | 63.3 | |
| CMMS probeTarget Model=GPT-OSS-20B2026.06 | 62.3 | |
| LLM-as-a-JudgeTarget Model=GPT-OSS-20B2026.06 | 61.1 | |
| LR probeTarget Model=GPT-OSS-20B2026.06 | 60.8 | |
| MMS probeTarget Model=GPT-OSS-20B2026.06 | 54.6 | |
| LAT probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 49.9 | |
| LAT probeTarget Model=GPT-OSS-20B2026.06 | 45.3 |