FETA's CLIP-MIL (Many-Shot Image-to-text)
Reported on 6 benchmarks across 2 tasks · 1 paper · 6 SOTA
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Computer Vision3 results
- R@1· 2022-09-08SOTA29
- R@10· 2022-09-08SOTA72.6
- R@5· 2022-09-08SOTA59.9
Natural Language Processing3 results
- R@1· 2022-09-08SOTA35.5
- R@10· 2022-09-08SOTA67
- R@5· 2022-09-08SOTA58.3