Video-to-text retrieval on MSRVTT 9k
49.7R@1AVIGATE
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| AVIGATEBackbone=ViT-B/32, Modality=Audiovisual2026.03 | 49.7 | 75.3 | 83.7 | 208.7 | |
| AVIGATE+Backbone=ViT-B/32, Modality=Audiovisual2026.03 | 49.5 | 75.5 | 85.4 | 210.4 | |
| SAVEBackbone=ViT-B/32, Modality=Audiovisual2026.03 | 48.9 | 78 | 86.5 | 213.4 | |
| DiscoVLABackbone=ViT-B/32, Modality=Vision-only2026.03 | 47.7 | 73.6 | 83.6 | 204.9 | |
| UCoFiABackbone=ViT-B/32, Modality=Vision-only2026.03 | 47.1 | 74.3 | 83 | 204.4 | |
| TEFALBackbone=ViT-B/32, Modality=Audiovisual2026.03 | 47.1 | 75.1 | 84.9 | 207.1 | |
| UATVRBackbone=ViT-B/32, Modality=Vision-only2026.03 | 46.9 | 73.8 | 83.8 | 204.5 | |
| X-CLIPBackbone=ViT-B/32, Modality=Vision-only2026.03 | 46.8 | 73.3 | 84 | 204.1 | |
| DiCoSABackbone=ViT-B/32, Modality=Vision-only2026.03 | 46.7 | 75.2 | 84.3 | 206.2 | |
| SAVE-hBackbone=ViT-B/32, Modality=Audiovisual2026.03 | 46.7 | 76 | 85.6 | 208.3 | |
| ProSTBackbone=ViT-B/32, Modality=Vision-only2026.03 | 46.3 | 74.2 | 83.2 | 203.7 | |
| PromptSwitchBackbone=ViT-B/32, Modality=Vision-only2026.03 | 46 | 74.3 | 84.8 | 205.1 | |
| TempMeBackbone=ViT-B/32, Modality=Vision-only2026.03 | 45.6 | 72.4 | 81.2 | 199.2 | |
| TS2-NetBackbone=ViT-B/32, Modality=Vision-only2026.03 | 45.3 | 74.1 | 83.7 | 203.1 | |
| EERCFBackbone=ViT-B/32, Modality=Vision-only2026.03 | 44.7 | 74.2 | 83.9 | 202.8 | |
| EclipSEBackbone=ViT-B/32, Modality=Audiovisual2026.03 | 44.7 | 71.3 | 82.8 | 198.8 | |
| X-PoolBackbone=ViT-B/32, Modality=Vision-only2026.03 | 44.4 | 73.3 | 84 | 201.7 | |
| DGLBackbone=ViT-B/32, Modality=Vision-only2026.03 | 43.5 | 70.5 | 80.7 | 194.7 | |
| CLIP4ClipBackbone=ViT-B/32, Modality=Vision-only2026.03 | 43.1 | 70.5 | 81.2 | 194.8 |