Image Classification on Five Datasets 16-shot
79.7AccuracySWIFT
Evaluation results
| Method | Links | |
|---|---|---|
| SWIFTLearning Paradigm=FT, Backbone=OpenCLIP ViT-B/32, Retrieval Augmentation=true2025.12 | 79.7 | |
| Fully Supervised w/ RALearning Paradigm=FT, Backbone=OpenCLIP ViT-B/32, Retrieval Augmentation=true2025.12 | 77.2 | |
| Fully SupervisedLearning Paradigm=FT, Backbone=OpenCLIP ViT-B/322025.12 | 76 | |
| SWATLearning Paradigm=FT, Backbone=OpenCLIP ViT-B/32, Retrieval Augmentation=true2025.12 | 74 | |
| DebiasPL on VLM (ours)Learning Paradigm=FT, Backbone=OpenCLIP ViT-B/322025.12 | 73.2 | |
| FixMatch on VLM (ours)Learning Paradigm=FT, Backbone=OpenCLIP ViT-B/322025.12 | 71.2 | |
| FS-FTLearning Paradigm=FT, Backbone=OpenCLIP ViT-B/322025.12 | 69.1 | |
| FineSSLLearning Paradigm=PL, Backbone=OpenCLIP ViT-B/322025.12 | 68.9 | |
| CLAPLearning Paradigm=AL, Backbone=OpenCLIP ViT-B/322025.12 | 65.5 | |
| Linear ProbingLearning Paradigm=AL, Backbone=OpenCLIP ViT-B/322025.12 | 64.7 | |
| CMLPLearning Paradigm=AL, Backbone=OpenCLIP ViT-B/322025.12 | 64.2 | |
| TaskRes(e)Learning Paradigm=AL, Backbone=OpenCLIP ViT-B/322025.12 | 62 | |
| CLIP-AdapterLearning Paradigm=AL, Backbone=OpenCLIP ViT-B/322025.12 | 60 | |
| CoOpLearning Paradigm=PL, Backbone=OpenCLIP ViT-B/322025.12 | 59.5 | |
| PLOTLearning Paradigm=PL, Backbone=OpenCLIP ViT-B/322025.12 | 59.3 | |
| Tip-Adapter (f)Learning Paradigm=AL, Backbone=OpenCLIP ViT-B/322025.12 | 58 | |
| FixMatch on VLMLearning Paradigm=FT, Backbone=OpenCLIP ViT-B/322025.12 | 57.2 | |
| DebiasPL on VLMLearning Paradigm=FT, Backbone=OpenCLIP ViT-B/322025.12 | 57.1 |