Text-to-Video Retrieval on S-MiT full (test val)
60.8Recall@1umt-l 25M
Evaluation results
| Method | Links | |
|---|---|---|
| umt-l 25MZero-shot=true2025.03 | 60.8 | |
| umt-l 17MZero-shot=true2025.03 | 60.7 | |
| umt-b 25MZero-shot=true2025.03 | 53.9 | |
| internvid-L-200MZero-shot=true2025.03 | 53.7 | |
| umt-b 17MZero-shot=true2025.03 | 53.5 | |
| videomamba-vm-25MZero-shot=true2025.03 | 53.2 | |
| umt-l 5MZero-shot=true2025.03 | 51.9 | |
| videomamba-vm-17MZero-shot=true2025.03 | 51.6 | |
| internvid-L-10M-FLTZero-shot=true2025.03 | 50.5 | |
| internvid-B-200MZero-shot=true2025.03 | 49.8 | |
| internvid-B-10M-FLTZero-shot=true2025.03 | 48.9 | |
| internvid-L-10M-DIVZero-shot=true2025.03 | 48.9 | |
| umt-b 5MZero-shot=true2025.03 | 47.9 | |
| videomamba-vm-5MZero-shot=true2025.03 | 47.6 | |
| internvid-L-50MZero-shot=true2025.03 | 45.7 | |
| internvid-L-WV10MZero-shot=true2025.03 | 38.8 | |
| internvid-L-10MZero-shot=true2025.03 | 35.9 |