Visual Question Answering on M3D 1.0 (test)
77.79AccuracyHulu-L1
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Hulu-L1Base Model=Hulu-Med-7B2026.03 | 77.79 | 47.334 | 47.167 | 45.234 | 12.491 | 66.03 | |
| VisionZipBase Model=Hulu-Med-7B2026.03 | 77.623 | 47.596 | 47.418 | 45.451 | 12.549 | 69.96 | |
| HiPruneBase Model=Hulu-Med-7B2026.03 | 77.616 | 47.072 | 46.199 | 45.417 | 12.574 | 22.3 | |
| MedPrunerBase Model=Hulu-Med-7B2026.03 | 77.452 | 47.434 | 47.262 | 45.304 | 12.58 | 52.1 | |
| Original (Hulu-Med-7B)Base Model=Hulu-Med-7B2026.03 | 75.317 | 34.188 | 34.047 | 31.82 | 9.541 | 100 | |
| VisionZipBase Model=MedGemma1.5-4B2026.03 | 45.428 | 7.925 | 4.364 | 2.707 | 0.639 | 69.98 | |
| Hulu-L1Base Model=MedGemma1.5-4B2026.03 | 44.638 | 7.81 | 4.473 | 2.779 | 0.651 | 65.98 | |
| HiPruneBase Model=MedGemma1.5-4B2026.03 | 44.42 | 8.201 | 4.447 | 2.792 | 0.664 | 22.3 | |
| MedPrunerBase Model=MedGemma1.5-4B2026.03 | 43.718 | 8.983 | 5.845 | 3.923 | 1.005 | 4.87 | |
| Original (MedGemma1.5-4B)Base Model=MedGemma1.5-4B2026.03 | 32.717 | 6.797 | 5.888 | 3.741 | 0.835 | 100 |