Image-to-Text retrieval on CREPE (test)
0.8735Multi-Obj AccCLIP + Linear
Evaluation Results
| Method | Links | |
|---|---|---|
| CLIP + LinearBase Model=CLIP, Adaptation Strategy=Linear2023.05 | 0.8735 | |
| CLIP + FT lateBase Model=CLIP, Adaptation Strategy=Late Fine-tuning2023.05 | 0.8714 | |
| MM-AdapterBase Model=CLIP, Adaptation Strategy=MM-Adapter (multimodal feature adapter)2023.05 | 0.8702 | |
| CLIP + FT allBase Model=CLIP, Adaptation Strategy=Full Fine-tuning2023.05 | 0.8239 | |
| MM-PredBase Model=FLAVA, Adaptation Strategy=MM-Pred (multimodal head)2023.05 | 0.8137 | |
| CLIP + Prompt-tuneBase Model=CLIP, Adaptation Strategy=Prompt-tuning2023.05 | 0.8081 | |
| MM-PredBase Model=CLIP, Adaptation Strategy=MM-Pred (multimodal head)2023.05 | 0.7784 | |
| CLIPBase Model=CLIP2023.05 | 0.7743 | |
| MM-AdapterBase Model=FLAVA, Adaptation Strategy=MM-Adapter (multimodal feature adapter)2023.05 | 0.7481 | |
| FLAVABase Model=FLAVA2023.05 | 0.651 | |
| FLAVA + FT lateBase Model=FLAVA, Adaptation Strategy=Late Fine-tuning2023.05 | 0.5811 | |
| FLAVA + LinearBase Model=FLAVA, Adaptation Strategy=Linear2023.05 | 0.551 | |
| Randomtype=baseline2023.05 | 0.5 |