Image, Text-to-Multimodal Retrieval on InfoSeek
48.9Recall@5CLIP-SF
Evaluation Results
| Method | Links | |
|---|---|---|
| CLIP-SFRetrieval Strategy=Embedding-based, Fusion Strategy=Score-level Fusion, Re-ranking=false, Retrieval Pool=Task-specific2025.03 | 48.9 | |
| CLIP-SFApproach=Embedding-based, Fusion=SF2025.03 | 44.6 | |
| BLIP-FFRetrieval Strategy=Embedding-based, Fusion Strategy=Feature-level Fusion, Re-ranking=false, Retrieval Pool=Task-specific2025.03 | 33 | |
| GENIUSRetrieval Strategy=Generative, Fusion Strategy=None, Re-ranking=true, Retrieval Pool=Task-specific2025.03 | 30.1 | |
| GENIUSApproach=Generative, Re-ranking=true2025.03 | 28.6 | |
| BLIP-FFApproach=Embedding-based, Fusion=FF2025.03 | 27.7 | |
| GENIUSRetrieval Strategy=Generative, Fusion Strategy=None, Re-ranking=false, Retrieval Pool=Task-specific2025.03 | 14.6 | |
| GENIUSApproach=Generative2025.03 | 14.2 |