Deception Detection on Insider Trading
100AUROCSelf-Evaluation
Evaluation Results
| Method | Links | |
|---|---|---|
| Self-EvaluationTarget Model=GEMMA4-26B-A4B-IT2026.06 | 100 | |
| STATEWITNESSTarget Model=GEMMA4-26B-A4B-IT2026.06 | 100 | |
| LLM-as-a-Judge w/ CoTTarget Model=GEMMA4-26B-A4B-IT2026.06 | 99.7 | |
| LLM-as-a-JudgeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 99.6 | |
| LAT probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 98.6 | |
| STATEWITNESSTarget Model=GPT-OSS-20B2026.06 | 98.2 | |
| LR probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 90.6 | |
| LLM-as-a-Judge w/ CoTTarget Model=GPT-OSS-20B2026.06 | 89.3 | |
| LLM-as-a-JudgeTarget Model=GPT-OSS-20B2026.06 | 87.9 | |
| MMS probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 82.7 | |
| Self-EvaluationTarget Model=GPT-OSS-20B2026.06 | 73.2 | |
| CMMS probeTarget Model=GPT-OSS-20B2026.06 | 67.2 | |
| CMMS probeTarget Model=GEMMA4-26B-A4B-IT2026.06 | 58 | |
| LAT probeTarget Model=GPT-OSS-20B2026.06 | 56.6 | |
| MMS probeTarget Model=GPT-OSS-20B2026.06 | 55 | |
| LR probeTarget Model=GPT-OSS-20B2026.06 | 46.6 |