Image-to-Text Retrieval on MSCOCO Caption
77R@1ALIGN
Evaluation Results
| Method | Links | |
|---|---|---|
| ALIGNtuning_mode=fine-tuning (FT100%), resolution=2892022.06 | 77 | |
| Uni-Perceiver-L + Conditional MoEstuning_mode=fine-tuning (FT100%)2022.06 | 74.7 | |
| Uni-Perceiver-Ltuning_mode=fine-tuning (FT100%)2022.06 | 74.4 | |
| Uni-Perceiver-Ltuning_mode=prompt tuning (PT1%)2022.06 | 73.3 | |
| Uni-Perceiver-L + Conditional MoEstuning_mode=prompt tuning (PT1%)2022.06 | 73.3 | |
| Uni-Perceiver-B + Conditional MoEstuning_mode=fine-tuning (FT100%)2022.06 | 70.5 | |
| Uni-Perceiver-Btuning_mode=fine-tuning (FT100%)2022.06 | 69.8 | |
| Uni-Perceiver-B + Conditional MoEstuning_mode=prompt tuning (PT1%)2022.06 | 68.9 | |
| Uni-Perceiver-Btuning_mode=prompt tuning (PT1%)2022.06 | 68.4 | |
| Uni-Perceiver-L + Conditional MoEstuning_mode=w/o tuning (WT)2022.06 | 67.9 | |
| Uni-Perceiver-Ltuning_mode=w/o tuning (WT)2022.06 | 67.8 | |
| ImageBERTtuning_mode=fine-tuning (FT100%)2022.06 | 66.4 | |
| CoCa-Ltuning_mode=w/o tuning (WT), resolution=5762022.06 | 66.3 | |
| Flamingo-3Btuning_mode=w/o tuning (WT), resolution=3202022.06 | 65.9 | |
| UNITER-Ltuning_mode=w/o tuning (WT)2022.06 | 65.7 | |
| Uni-Perceiver-Btuning_mode=w/o tuning (WT)2022.06 | 64.9 | |
| Florencetuning_mode=w/o tuning (WT), resolution=3842022.06 | 64.7 | |
| Uni-Perceiver-B + Conditional MoEstuning_mode=w/o tuning (WT)2022.06 | 64.6 | |
| UNITER-Btuning_mode=w/o tuning (WT)2022.06 | 64.4 | |
| ViLTtuning_mode=fine-tuning (FT100%)2022.06 | 64.4 | |
| CoCa-Btuning_mode=w/o tuning (WT), resolution=5762022.06 | 63.8 | |
| ALIGNtuning_mode=w/o tuning (WT), resolution=2892022.06 | 58.6 | |
| CLIPtuning_mode=w/o tuning (WT), resolution=3362022.06 | 58.4 | |
| ViLTtuning_mode=w/o tuning (WT)2022.06 | 55 | |
| ImageBERTtuning_mode=w/o tuning (WT)2022.06 | 44 | |
| FLAVAtuning_mode=w/o tuning (WT)2022.06 | 42.7 |