Overall Hallucination Detection (E3+E4) on 50 Patients (test)
82.3F1 ScoreQwen3-14B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen3-14BType=Curated FT2026.05 | 82.3 | 83.5 | 81.2 | |
| Qwen3-14BType=Base2026.05 | 77.7 | 74.9 | 80.8 | |
| Qwen3-14BType=Original FT2026.05 | 75.3 | 84.1 | 68.1 | |
| Qwen3-PlusType=API2026.05 | 65.8 | 83.3 | 54.4 |