Medical Image Classification on OrganAMNIST
93.4AccuracyVISReg
Evaluation results
| Method | Links | |
|---|---|---|
| VISRegBackbone=ViT-B/16, Heuristics=Excluded, Evaluation protocol=Linear probe2026.06 | 93.4 | |
| VISReg-Inet22KBackbone=ViT-L/14, Training Scale=large scale datasets, Evaluation protocol=Linear probe2026.06 | 93.33 | |
| VISRegBackbone=ViT-L/14, Heuristics=Excluded, Evaluation protocol=Linear probe2026.06 | 92.93 | |
| I-JEPABackbone=ViT-H/14, Heuristics=Included, Evaluation protocol=Linear probe2026.06 | 92.19 | |
| iBOTBackbone=ViT-B/16, Heuristics=Included, Evaluation protocol=Linear probe2026.06 | 91.84 | |
| DINOBackbone=ViT-B/16, Heuristics=Included, Evaluation protocol=Linear probe2026.06 | 91.7 | |
| MoCoV3Backbone=ViT-B/16, Heuristics=Included, Evaluation protocol=Linear probe2026.06 | 91.37 | |
| DINOv2-LVD142MBackbone=ViT-L/14, Training Scale=large scale datasets, Evaluation protocol=Linear probe2026.06 | 91.01 | |
| iBOTBackbone=ViT-L/16, Heuristics=Included, Evaluation protocol=Linear probe2026.06 | 90.82 | |
| MAEBackbone=ViT-L/16, Heuristics=Excluded, Evaluation protocol=Linear probe2026.06 | 90.06 | |
| data2vecBackbone=ViT-L/16, Heuristics=Included, Evaluation protocol=Linear probe2026.06 | 89.04 | |
| BioVLMTrain Params=30K2026.04 | 54.75 | |
| PromptSRCTrain Params=69K2026.04 | 47.27 | |
| CoCoOpTrain Params=44K2026.04 | 46.39 | |
| KgCoOpTrain Params=3K2026.04 | 45.72 | |
| TCPTrain Params=0.5M2026.04 | 43.23 | |
| CoOpTrain Params=3K2026.04 | 39.97 | |
| MaPLeTrain Params=5.3M2026.04 | 39.46 | |
| BiomedCoOpTrain Params=3K2026.04 | 36.6 | |
| ProGradTrain Params=3K2026.04 | 33.93 | |
| BioMedCLIPTrain Params=02026.04 | 24.99 |