Zero Shot Transfer Image Classification On 4
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Accuracy | Paper Title | Repository |
|---|---|---|---|
| CoCa | 96.5 | CoCa: Contrastive Captioners are Image-Text Foundation Models | - |
| CLIP | 88.9 | Learning Transferable Visual Models From Natural Language Supervision | - |
| ALIGN | 92.2 | Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision | - |
| BASIC (Lion) | 96.8 | - | - |
| PaLI | 81.97 | PaLI: A Jointly-Scaled Multilingual Language-Image Model | - |
| EVA-CLIP-E/14+ | 94.5 | EVA-CLIP: Improved Training Techniques for CLIP at Scale | - |
| LiT ViT-e | 96.1 | PaLI: A Jointly-Scaled Multilingual Language-Image Model | - |
| BASIC | 95.7 | Combined Scaling for Zero-shot Transfer Learning | - |
| LiT-22B | 96.0 | Scaling Vision Transformers to 22 Billion Parameters | - |
| AltCLIP | 87.2 | AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities | - |
| EVA-CLIP-18B | 95.7 | EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters | - |
| LiT-tuning | 93.9 | LiT: Zero-Shot Transfer with Locked-image text Tuning | - |
0 of 12 row(s) selected.