Domain Generalization On Terraincognita

评估指标

Average Accuracy

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
UniDG + CORAL + ConvNeXt-B	69.6	Towards Unified and Effective Domain Generalization
GMDG (RegNetY-16GF, SWAD)	65	Rethinking Multi-domain Generalization with A General Learning Objective
MIRO (RegNetY-16GF, SWAD)	64.3	Domain Generalization by Mutual-Information Regularization with Pre-trained Models
CAR-FT (CLIP, ViT-B/16)	61.9	Context-Aware Robust Fine-Tuning	-
Ensemble of Averages (RegNetY-16GF)	61.1	Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization
GMDG (RegNetY-16GF)	60.7	Rethinking Multi-domain Generalization with A General Learning Objective
SIMPLE+	59.0	SIMPLE: Specialized Model-Sample Matching for Domain Generalization	-
VL2V-SD (CLIP, ViT-B/16)	58.54	Leveraging Vision-Language Models for Improving Domain Generalization in Image Classification
SIMPLE	57.6	SIMPLE: Specialized Model-Sample Matching for Domain Generalization	-
SEDGE+	56.8	Domain Generalization using Pretrained Models without Fine-tuning	-
SEDGE	56.8	Domain Generalization using Pretrained Models without Fine-tuning	-
CADG	55.7	CADG: A Model Based on Cross Attention for Domain Generalization	-
Ensemble of Averages (ResNeXt-50 32x4d)	55.2	Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization
EOQ (ResNet-50)	53.2	QT-DoG: Quantization-aware Training for Domain Generalization
GMDG (ResNet-50, SWAD)	53.0	Rethinking Multi-domain Generalization with A General Learning Objective
MIRO (ResNet-50, SWAD)	52.9	Domain Generalization by Mutual-Information Regularization with Pre-trained Models
MoA (OpenCLIP, ViT-B/16)	52.8	Domain Generalization Using Large Pretrained Models with Mixture-of-Adapters
Ensemble of Averages (ResNet-50)	52.3	Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization
Model Ratatouille	52	Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization
VNE (ResNet-50, SWAD)	51.7	VNE: An Effective Method for Improving Deep Representation by Manipulating Eigenvalue Distribution

0 of 30 row(s) selected.