Forgery Detection on MedForge Real Forgery (In-Domain) 90K
99.24AccuracyMedForge-Reasoner
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MedForge-ReasonerCategory=Generic MLLMs2026.03 | 99.24 | 98.86 | |
| MedForge-ReasonerCategory=Generic MLLMs2026.03 | 99.24 | 98.86 | |
| SIDA-13BCategory=Specialized Detectors2026.03 | 91.52 | 90.86 | |
| AIGC-HolmesCategory=Specialized Detectors2026.03 | 90.24 | 88.57 | |
| SIDA-13BCategory=Specialized Detectors2026.03 | 89.37 | 88.69 | |
| FakeVLMCategory=Specialized Detectors2026.03 | 88.93 | 88.27 | |
| AIGC-HolmesCategory=Specialized Detectors2026.03 | 88.12 | 90.46 | |
| FakeVLMCategory=Specialized Detectors2026.03 | 86.48 | 85.73 | |
| SIDA-7BCategory=Specialized Detectors2026.03 | 79.83 | 86.47 | |
| SIDA-7BCategory=Specialized Detectors2026.03 | 77.56 | 76.84 | |
| Gemini3-ProCategory=Generic MLLMs2026.03 | 75.34 | 74.22 | |
| Gemini3-FlashCategory=Generic MLLMs2026.03 | 71.26 | 24.42 | |
| Gemini3-ProCategory=Generic MLLMs2026.03 | 67.29 | 61.82 | |
| Qwen3VL-FlashCategory=Generic MLLMs2026.03 | 57.6 | 32.48 | |
| Qwen3VL-PlusCategory=Generic MLLMs2026.03 | 57.42 | 54.7 | |
| Gemini3-FlashCategory=Generic MLLMs2026.03 | 57.33 | 70.22 | |
| Qwen3VL-PlusCategory=Generic MLLMs2026.03 | 54.14 | 47.73 | |
| Qwen3VL-FlashCategory=Generic MLLMs2026.03 | 47.83 | 55.08 |