Image Classification on DMLAB
18.7Top-1 AccIL-CLIP
Evaluation results
| Method | Links | |
|---|---|---|
| IL-CLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 18.7 | |
| NegCLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 15.9 | |
| IL-CLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 15.3 | |
| IL-CLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.04 | 15 | |
| NegCLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.04 | 14 | |
| CLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 13.6 | |
| CLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.04 | 13 | |
| Codebook-CLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.04 | 13 | |
| NegCLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 11.9 | |
| CLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 11.7 | |
| Codebook-CLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 11.7 | |
| Codebook-CLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.04 | 4.8 |