List your company
ResearchFormerly WizWand

Image Classification on DermaMNIST

86.2AccuracyBiomedGPT
1.54423.52245.567.478Oct 13, 2022May 25, 2023Jan 5, 2024Aug 17, 2024Mar 30, 2025Nov 10, 2025Jun 23, 2026
Updated 4mo ago

Evaluation results

MethodLinks
BiomedGPTCategory=Finetuned Medical MLLMs2026.0386.2
Clinician Mimetic WorkflowCategory=Proposed2026.0383.6
SigLIP2-BCategory=Supervised (Full-Tuning)2026.0381.8
SaEBudget=20%, Number of seeds=52026.0280.21
Swin-BCategory=Supervised (Full-Tuning)2026.0379.5
Hulu-MedCategory=Finetuned Medical MLLMs2026.0377.7
ViT-BCategory=Supervised (Full-Tuning)2026.0376.6
L1 KDE-XEBackbone=ResNet-110 (SD)2022.1076.4
TIESBackbone=ViT-L/16, Number of samples=5002026.0676.4
ConvNeXt-BCategory=Supervised (Full-Tuning)2026.0376.2
DARE-TIESBackbone=ViT-L/16, Number of samples=5002026.0676.2
Domain onlyBackbone=ViT-L/16, Number of samples=5002026.0676.1
TaDABackbone=ViT-L/16, Number of samples=5002026.0676
MMCEBackbone=DenseNet-402022.1075.8
SVD MergeBackbone=ViT-L/16, Number of samples=5002026.0675.8
MedCoOp + BADGEBudget=20%, Number of seeds=52026.0275.46
L1 KDE-XEBackbone=Wide-ResNet-28-102022.1075.4
MMCEBackbone=ResNet-110 (SD)2022.1075.3
LinearBackbone=ViT-L/16, Number of samples=5002026.0675
L1 KDE-XEBackbone=DenseNet-402022.1074.8
Task onlyBackbone=ViT-L/16, Number of samples=5002026.0674.8
MedCoOp + EntropyBudget=20%, Number of seeds=52026.0274.56
L1 KDE-XEBackbone=ResNet-1102022.1074.4
XEBackbone=ResNet-110 (SD)2022.1074.3
DARE-LinBackbone=ViT-L/16, Number of samples=5002026.0674.2
Mag. PruningBackbone=ViT-L/16, Number of samples=5002026.0674.2
MedCoOp + CoresetBudget=20%, Number of seeds=52026.0274.11
MMCEBackbone=Wide-ResNet-28-102022.1074.1
XEBackbone=DenseNet-402022.1074.1
XEBackbone=Wide-ResNet-28-102022.1073.6
ResNet50Category=Supervised (Full-Tuning)2026.0373
Task Arith.Backbone=ViT-L/16, Number of samples=5002026.0672.6
Base modelBackbone=ViT-L/16, Number of samples=5002026.0672.2
MMCEBackbone=ResNet-1102022.1072.1
XEBackbone=ResNet-1102022.1072
FL-53Backbone=Wide-ResNet-28-102022.1071.5
PCBBudget=20%, Number of seeds=52026.0271.07
FL-53Backbone=DenseNet-402022.1070.5
RandomBudget=20%, Number of seeds=52026.0269.42
FL-53Backbone=ResNet-110 (SD)2022.1068.9
FL-53Backbone=ResNet-1102022.1067.4
BioMedVRSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0665.3
VPSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0664.9
BiomedCoOpBudget=20%, Number of seeds=5, inference_mode=few-shot reference2026.0262.59
Gemini-3-flashCategory=General MLLMs2026.0362.1
AttrVRSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0661.6
BioMedCoOpSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0659.9
ARSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0658
BiomedCLIPPrediction Mode=Zero-shot2026.0345.89
CoCoOpSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0644
CoOpSetting=Few-shot, Shots=16, Backbone=ViT-B/162026.0643
GPT-5-miniCategory=General MLLMs2026.0337.4
MedGemmaCategory=Medical MLLMs2026.0332.8
CLIPPrediction Mode=Zero-shot2026.0324.74
BioMedCLIPSetting=Zero-shot, Shots=0, Backbone=ViT-B/162026.0619.3
BioMedVRSetting=Zero-shot, Shots=0, Backbone=ViT-B/162026.0613.9
CLIPSetting=Zero-shot, Shots=0, Backbone=ViT-B/162026.0613
LLaVA-MedCategory=Medical MLLMs2026.0312.9
SigLIP2Prediction Mode=Zero-shot2026.0311.67
AttrVRSetting=Zero-shot, Shots=0, Backbone=ViT-B/162026.069.7
SigLIPPrediction Mode=Zero-shot2026.038.28
BioMedVRSetting=Zero-shot, Shots=0, Backbone=ViT-B/16, Pre-trained Model=BioMedCLIP2026.064.8