Hallucination Detection on Clinical Hallucination E4-level (test)
0.831F1 Score (E4)Qwen3-14B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen3-14BModel Configuration=Curated FT2026.05 | 0.831 | 0.909 | 0.765 | |
| Qwen3-14BModel Configuration=Original FT2026.05 | 0.803 | 0.748 | 0.867 | |
| Qwen3-PlusModel Configuration=API2026.05 | 0.791 | 0.745 | 0.842 | |
| Qwen3-14BModel Configuration=Base2026.05 | 0.554 | 0.43 | 0.779 |