Image Classification on PathMNIST
96.65AccuracyHamCls
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HamClsResolution=224x2242026.03 | 96.65 | 99.36 | |
| MedMamba-BResolution=224x2242026.03 | 96.4 | 99.9 | |
| BiomedGPTCategory=Finetuned Medical MLLMs2026.03 | 95.6 | — | |
| Clinician Mimetic WorkflowCategory=Proposed2026.03 | 95.4 | — | |
| MedViT-SResolution=224x2242026.03 | 94.2 | 99.3 | |
| SigLIP2-BCategory=Supervised (Full-Tuning)2026.03 | 94.1 | — | |
| Domain onlyBackbone=ViT-L/16, Number of samples=5002026.06 | 93.4 | — | |
| Hulu-MedCategory=Finetuned Medical MLLMs2026.03 | 92.6 | — | |
| Swin-BCategory=Supervised (Full-Tuning)2026.03 | 92 | — | |
| TaDABackbone=ViT-L/16, Number of samples=5002026.06 | 92 | — | |
| DARE-TIESBackbone=ViT-L/16, Number of samples=5002026.06 | 91.8 | — | |
| ConvNeXt-BCategory=Supervised (Full-Tuning)2026.03 | 91.1 | — | |
| LinearBackbone=ViT-L/16, Number of samples=5002026.06 | 91.1 | — | |
| ViT-BCategory=Supervised (Full-Tuning)2026.03 | 90.9 | — | |
| Mag. PruningBackbone=ViT-L/16, Number of samples=5002026.06 | 90.8 | — | |
| DARE-LinBackbone=ViT-L/16, Number of samples=5002026.06 | 90.6 | — | |
| TIESBackbone=ViT-L/16, Number of samples=5002026.06 | 90.2 | — | |
| SVD MergeBackbone=ViT-L/16, Number of samples=5002026.06 | 89.9 | — | |
| ResNet50Category=Supervised (Full-Tuning)2026.03 | 89.6 | — | |
| Task Arith.Backbone=ViT-L/16, Number of samples=5002026.06 | 89.4 | — | |
| GIF-SDBackbone=ResNet-50, Expansion Factor=5x2022.11 | 86.9 | — | |
| GIF-SDBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 86.9 | — | |
| Base modelBackbone=ViT-L/16, Number of samples=5002026.06 | 86.7 | — | |
| Task onlyBackbone=ViT-L/16, Number of samples=5002026.06 | 86.6 | — | |
| SDBackbone=ResNet-50, Expansion Factor=5x2022.11 | 85.1 | — | |
| SDBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 85.1 | — | |
| GIF-DALLEBackbone=ResNet-50, Expansion Factor=5x2022.11 | 84.4 | — | |
| GIF-DALLEBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 84.4 | — | |
| DALL-E2Backbone=ResNet-50, Expansion Factor=5x2022.11 | 82.8 | — | |
| DALL-E2Backbone=ResNet-50, Expansion Ratio=5x2022.11 | 82.8 | — | |
| GIF-MAEBackbone=ResNet-50, Expansion Factor=5x2022.11 | 82 | — | |
| GIF-MAEBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 82 | — | |
| MAEBackbone=ResNet-50, Expansion Factor=5x2022.11 | 81.7 | — | |
| MAEBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 81.7 | — | |
| RandAugmentBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 79.2 | — | |
| CutoutBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 78.8 | — | |
| GridMaskBackbone=ResNet-50, Expansion Ratio=5x2022.11 | 78.4 | — | |
| Distillation of CLIPBackbone=ResNet-502022.11 | 77.3 | — | |
| OriginalBackbone=ResNet-50, Expansion Factor=1x2022.11 | 72.4 | — | |
| OriginalBackbone=ResNet-50, Expansion Ratio=1x2022.11 | 72.4 | — | |
| Gemini-3-flashCategory=General MLLMs2026.03 | 64.9 | — | |
| MedGemmaCategory=Medical MLLMs2026.03 | 43.2 | — | |
| GPT-5-miniCategory=General MLLMs2026.03 | 39.3 | — | |
| LLaVA-MedCategory=Medical MLLMs2026.03 | 16.4 | — | |
| CLIPBackbone=CLIP2022.11 | 10.7 | — |