CompoDiff (CLIP L/14)
Reported on 6 benchmarks across 2 tasks · 1 paper
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Computer Vision6 results
- (Recall@10+Recall@50)/2· 2023-03-2144.11best: 71.77 (DQU-CIR)
- mAP@10· 2023-03-2113.51best: 43.4 (MMRet-MLLM)
- R@5· 2023-03-2154.36best: 81.08 (CoVR-BLIP-2)
- (Recall@10+Recall@50)/2· 2023-03-2144.11best: 60.57 (CoVR-BLIP-2)
- mAP@10· 2023-03-2113.51best: 43.4 (MMRet-MLLM)
- R@5· 2023-03-2154.36best: 81.08 (CoVR-BLIP-2)