Image Retrieval On Wit
Metriken
R@1
R@5
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | R@1 | R@5 | Paper Title | Repository |
|---|---|---|---|---|
| WIT-ALL | 0.346 | 0.642 | WIT: Wikipedia-based Image Text Dataset for Multimodal Multilingual Machine Learning | |
| CC (Conceptual Captions) | 0.048 | 0.122 | WIT: Wikipedia-based Image Text Dataset for Multimodal Multilingual Machine Learning |
0 of 2 row(s) selected.