Medical Visual Question Answering on VQA-RAD (ECE, AUC)
16.85ECEMS-FBI
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MS-FBIModel=MedVLM-R12026.06 | 16.85 | 59.03 | |
| MS-FBIModel=Llava-NeXT-7B2026.06 | 20.9 | 56.88 | |
| MS-FBIModel=Llava-1.5-med-7B2026.06 | 21.77 | 56.8 | |
| MS-FBIModel=Molmo-7B2026.06 | 24.52 | 52.39 | |
| VanillaModel=MedVLM-R12026.06 | 25.59 | 54.95 | |
| PunishModel=MedVLM-R12026.06 | 27.63 | 57.59 | |
| Top-KModel=MedVLM-R12026.06 | 28.91 | 59.29 | |
| VanillaModel=Molmo-7B2026.06 | 30.28 | 49.87 | |
| Top-KModel=Molmo-7B2026.06 | 31.04 | 56.47 | |
| PunishModel=Molmo-7B2026.06 | 32.47 | 46.98 | |
| Top-KModel=Llava-NeXT-7B2026.06 | 33.69 | 54.03 | |
| VanillaModel=Llava-NeXT-7B2026.06 | 35.16 | 54.56 | |
| PunishModel=Llava-NeXT-7B2026.06 | 36.37 | 55.85 | |
| PunishModel=Llava-1.5-med-7B2026.06 | 38.31 | 51.3 | |
| VanillaModel=Llava-1.5-med-7B2026.06 | 39.44 | 50.63 | |
| Top-KModel=Llava-1.5-med-7B2026.06 | 42.41 | 50 |