Binary Alzheimer's Disease Classification (CN vs. AD) on OASIS (test)
86.41AUCMVSC (I-JEPA)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MVSC (I-JEPA)Backbone=ViT-H/14, Compression strategy=MVSC2026.01 | 86.41 | 83.85 | |
| MERLINMode=image encoder only2026.01 | 85.88 | 85.09 | |
| ViTBackbone=ViT-S/162026.01 | 85.47 | 83.58 | |
| MVSC (DINOv3)Backbone=ViT-B/16, Compression strategy=MVSC2026.01 | 84.79 | 83.54 | |
| Mean (I-JEPA)Backbone=ViT-H/14, Compression strategy=Mean pooling2026.01 | 80.87 | 85.01 | |
| 3D ResNetBackbone=ResNet502026.01 | 80.22 | 76.71 | |
| Max (I-JEPA)Backbone=ViT-H/14, Compression strategy=Max pooling2026.01 | 79.14 | 83.22 | |
| RAPTOR2026.01 | 77.72 | 80.12 | |
| PPAL2026.01 | 72.35 | 85.71 | |
| M3T2026.01 | 70.1 | 77.09 | |
| Mean (DINOv3)Backbone=ViT-B/16, Compression strategy=Mean pooling2026.01 | 69.54 | 73.6 | |
| AXIAL2026.01 | 57.61 | 65.15 | |
| Max (DINOv3)Backbone=ViT-B/16, Compression strategy=Max pooling2026.01 | 56.91 | 73.91 |