List your company
ResearchFormerly WizWand

Image Classification on DomainNet

90.32AccuracyPEP-FedPT
1.76424.754547.74570.7355Oct 18, 2021Jun 30, 2022Mar 13, 2023Nov 23, 2023Aug 5, 2024Apr 17, 2025Dec 29, 2025
Updated 5mo ago

Evaluation results

MethodLinks
PEP-FedPT2025.1090.32
pFedPG2025.1088.34
Fed-VPT-D2025.1087.08
SGPT2025.1086.55
FedPR2025.1083.59
Fed-VPT2025.1082.92
P-PT2025.1082.89
Head2025.1082.34
Full FTModel=ViT-Base2025.0573.46
LoFTModel=ViT-Base, Rank=162025.0571.97
LoRAModel=ViT-Base, Rank=162025.0571.39
Manually AnnotatedType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0371.13
DoRAModel=ViT-Base, Rank=162025.0570.18
LoFTModel=ViT-Base, Rank=82025.0570.11
LoFTModel=ViT-Base, Rank=42025.0569.32
Any-shift Prompting + TPTTest-time optimization=true, Backbone=ViT-B/162024.0261.75
Any-shift PromptingTest-time optimization=false, Backbone=ViT-B/162024.0260.93
CoOp + TPTTest-time optimization=true, Backbone=ViT-B/162024.0260.65
CoCoOp + TPTTest-time optimization=true, Backbone=ViT-B/162024.0260.44
TPTTest-time optimization=true, Backbone=ViT-B/162024.0259.9
DPLTest-time optimization=false, Backbone=ViT-B/162024.0259.86
CoCoOpTest-time optimization=false, Backbone=ViT-B/162024.0259.43
CoOpTest-time optimization=false, Backbone=ViT-B/162024.0258.82
CLIP-DTest-time optimization=false, Backbone=ViT-B/162024.0256.24
HIRPG (+) CONANType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0354.85
CLIPTest-time optimization=false, Backbone=ViT-B/162024.0254.08
LE (+) CONANType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0352.87
CHB (+) CONANType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0352.31
HIRPGType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0352.3
Noun SubmanifoldBackbone=CLIP ViT-B-32, Evaluation Protocol=Zero-shot2023.0550.8
CLIPBackbone=CLIP ViT-B-32, Evaluation Protocol=Zero-shot2023.0550.6
POS PCABackbone=CLIP ViT-B-32, Evaluation Protocol=Zero-shot2023.0550.6
POS PGABackbone=CLIP ViT-B-32, Evaluation Protocol=Zero-shot2023.0550.4
SC (+) CONANType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0349.64
GMOE-S/16Backbone=ViT-S/16, Selection Criterion=train-validation selection2022.0648.7
TSDBackbone=ResNet-502023.0347.73
LEType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0347.43
ERMBackbone=ViT-S/16, Selection Criterion=train-validation selection2022.0647.3
SWADBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0646.5
SHOT-IMBackbone=ResNet-502023.0346.3
T3ABackbone=ResNet-502023.0346.16
ETABackbone=ResNet-502023.0346.13
CHBType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0345.28
ERMBackbone=ResNet-502023.0345.2
EVIL-SAMOptimization=SAM2025.1245
TentBackbone=ResNet-502023.0344.71
MIROBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0644.3
BNBackbone=ResNet-502023.0343.38
LAMEBackbone=ResNet-502023.0343.2
SparseIRM-SAMOptimization=SAM2025.1243.1
CCG (+) CONANType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0342.94
mDSDIBackbone=ResNet-502021.1042.8
FISHBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0642.7
CORAL-SAMOptimization=SAM2025.1242.1
DANNBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0641.8
CORALBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0641.8
FishrBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0641.7
CORALBackbone=ResNet-502021.1041.5
MLDGBackbone=ResNet-502021.1041.2
ERMBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0641.2
PLBackbone=ResNet-502023.0341.18
ERMBackbone=ResNet-502021.1040.9
MTLBackbone=ResNet-502021.1040.6
MRM-SAMOptimization=SAM2025.1240.6
SCType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0340.42
SagNetsBackbone=ResNet-502021.1040.3
TSDBackbone=ResNet-182023.0340.19
SagNet-SAMOptimization=SAM2025.1240
T3ABackbone=ResNet-182023.0339.61
SHOT-IMBackbone=ResNet-182023.0339.5
ETABackbone=ResNet-182023.0339.43
MMDBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0639.4
CCGType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0339.32
MixupBackbone=ResNet-502021.1039.2
ERMBackbone=ResNet-182023.0338.95
TentBackbone=ResNet-182023.0338.95
RSCBackbone=ResNet-502021.1038.9
DANNBackbone=ResNet-502021.1038.3
CDANNBackbone=ResNet-502021.1038.3
Glide-synType=Training-dependent, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0338.26
BNBackbone=ResNet-182023.0337.8
LAMEBackbone=ResNet-182023.0337.49
ARMBackbone=ResNet-502021.1035.5
PLBackbone=ResNet-182023.0335.23
IRMBackbone=ResNet-502021.1033.9
IRMBackbone=ResNet-50, Selection Criterion=train-validation selection2022.0633.9
VREXBackbone=ResNet-502021.1033.6
GroupDROBackbone=ResNet-502021.1033.3
MMDBackbone=ResNet-502021.1023.4
SuS-X-SDType=Training-free, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0320.06
CuPLType=Training-free, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0318.25
VisDescType=Training-free, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0316.87
CALIPType=Training-free, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0316.43
SD-ClfType=Training-free, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.0312.27
CLIP-ZSType=Training-free, Vision Encoder=ResNet50, Pre-training=YFCC100M-CLIP, Setup=Joint training2024.035.17