| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Bidirectional Retrieval | SARVLM-1M (test) | Mean Recall30.94 | 25 | |
| Text-to-Image Retrieval | SARVLM-1M (test) | R@113.58 | 25 | |
| Image-to-Text Retrieval | SARVLM-1M (test) | R@112.66 | 25 | |
| Image Captioning | SARVLM-1M (test) | BLEU-126.1 | 4 |