Delusion Presence Detection on ADAPT full N=122 (test)
87.3Micro F1GPT-OSS
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-OSSPrompt Level=Level 4 (Full Prompt)2026.05 | 87.3 | |
| Majority VotePrompt Level=Level 4 (Full Prompt)2026.05 | 87.2 | |
| Direct JudgePrompt Level=Level 4 (Full Prompt)2026.05 | 86.6 | |
| GLM-4.6Prompt Level=Level 1 (Category)2026.05 | 86.5 | |
| GLM-4.6Prompt Level=Level 4 (Full Prompt)2026.05 | 85.7 | |
| GPT-OSSPrompt Level=Level 1 (Category)2026.05 | 85.5 | |
| Majority VotePrompt Level=Level 1 (Category)2026.05 | 85.3 | |
| Qwen-3Prompt Level=Level 4 (Full Prompt)2026.05 | 83.9 | |
| DebatePrompt Level=Level 4 (Full Prompt)2026.05 | 83.5 | |
| Qwen-3Prompt Level=Level 1 (Category)2026.05 | 80.9 |