Text-to-Video Retrieval on MSRVTT full (test val)
43.6Recall@1umt-l 17M
Evaluation Results
| Method | Links | |
|---|---|---|
| umt-l 17MBackbone=Large, Pretrain Data=17M, Zero-shot=true2025.03 | 43.6 | |
| umt-L-17MZero-shot=true, Re-ranking=true2025.03 | 43.6 | |
| umt-l 25MBackbone=Large, Pretrain Data=25M, Zero-shot=true2025.03 | 42.3 | |
| umt-L-25MZero-shot=true, Re-ranking=true2025.03 | 42.3 | |
| internvid-L-10M-FLTBackbone=Large, Pretrain Data=10M-FLT, Zero-shot=true2025.03 | 38.7 | |
| internvid-L-10M-FLTZero-shot=true, Re-ranking=true2025.03 | 38.7 | |
| internvid-L-200MBackbone=Large, Pretrain Data=200M, Zero-shot=true2025.03 | 38.2 | |
| internvid-L-200MZero-shot=true, Re-ranking=true2025.03 | 38.2 | |
| internvid-B-200MBackbone=Base, Pretrain Data=200M, Zero-shot=true2025.03 | 38.1 | |
| internvid-B-200MZero-shot=true, Re-ranking=true2025.03 | 38.1 | |
| internvid-B-10M-FLTBackbone=Base, Pretrain Data=10M-FLT, Zero-shot=true2025.03 | 37.9 | |
| internvid-B-10M-FLTZero-shot=true, Re-ranking=true2025.03 | 37.9 | |
| internvid-L-10M-DIVBackbone=Large, Pretrain Data=10M-DIV, Zero-shot=true2025.03 | 37.3 | |
| internvid-L-10M-DIVZero-shot=true, Re-ranking=true2025.03 | 37.3 | |
| umt-b 17MBackbone=Base, Pretrain Data=17M, Zero-shot=true2025.03 | 35.6 | |
| umt-B-17MZero-shot=true, Re-ranking=true2025.03 | 35.6 | |
| umt-b 25MBackbone=Base, Pretrain Data=25M, Zero-shot=true2025.03 | 35.3 | |
| umt-B-25MZero-shot=true, Re-ranking=true2025.03 | 35.3 | |
| videomamba-vm-17MVariant=VM, Pretrain Data=17M, Zero-shot=true2025.03 | 34.9 | |
| videomamba-vm-25MVariant=VM, Pretrain Data=25M, Zero-shot=true2025.03 | 34.9 | |
| videomamba-vm-17MZero-shot=true, Re-ranking=true2025.03 | 34.9 | |
| videomamba-vm-25MZero-shot=true, Re-ranking=true2025.03 | 34.9 | |
| umt-l 5MBackbone=Large, Pretrain Data=5M, Zero-shot=true2025.03 | 34.8 | |
| umt-L-5MZero-shot=true, Re-ranking=true2025.03 | 34.8 | |
| videomamba-vm-5MVariant=VM, Pretrain Data=5M, Zero-shot=true2025.03 | 33.3 | |
| videomamba-vm-5MZero-shot=true, Re-ranking=true2025.03 | 33.3 | |
| internvid-L-50MBackbone=Large, Pretrain Data=50M, Zero-shot=true2025.03 | 32.4 | |
| internvid-L-50MZero-shot=true, Re-ranking=true2025.03 | 32.4 | |
| umt-b 5MBackbone=Base, Pretrain Data=5M, Zero-shot=true2025.03 | 30 | |
| umt-B-5MZero-shot=true, Re-ranking=true2025.03 | 30 | |
| internvid-L-10MBackbone=Large, Pretrain Data=10M, Zero-shot=true2025.03 | 26.7 | |
| internvid-L-10MZero-shot=true, Re-ranking=true2025.03 | 26.7 | |
| internvid-L-WV10MBackbone=Large, Pretrain Data=WV10M, Zero-shot=true2025.03 | 26.5 | |
| internvid-L-WebVid10MZero-shot=true, Re-ranking=true2025.03 | 26.5 |