Video-to-Text Retrieval on MSRVTT (MdR, MnR)
77.1R@5EagleNet
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| EagleNetEnd-to-End Status=End-to-End, Backbone=ViT-B/162026.03 | 77.1 | 49.2 | 86.6 | 2 | 8.2 | |
| RAPEnd-to-End Status=End-to-End, Backbone=ViT-B/162026.03 | 76.4 | 45.3 | 84.8 | — | 9.1 | |
| TMASS†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 75.9 | 46.7 | 85.6 | 2 | 9.1 | |
| GLSCL*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 75.8 | 48.3 | 84.6 | 2 | 8.1 | |
| UCoFiA*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 75.7 | 49 | 84.2 | 2 | 10 | |
| TS2Net*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 75.4 | 46.4 | 85.3 | 2 | 8.6 | |
| TFVLEnd-to-End Status=End-to-End, Backbone=ViT-B/162026.03 | 75.3 | 47.3 | 83.7 | — | 9.5 | |
| ProST*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 75.2 | 48.3 | 84.5 | 2 | 8.7 | |
| TempMe*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 75.2 | 47.5 | 85.1 | 2 | 8.8 | |
| XPool†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 74.9 | 48 | 82.9 | 2 | 9.7 | |
| SHE-NetEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 74.8 | 44.8 | 83.7 | 2 | 9.1 | |
| CLIP-VIP*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 74.8 | 47.6 | 85 | 2 | 9.6 | |
| TV-ProxyNet†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 74.8 | 47.9 | 84.1 | 2 | 9.4 | |
| Video-ColBERT†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 74.4 | 47.9 | 84.9 | 2 | 9 | |
| EagleNetEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 73.7 | 45.7 | 83.4 | 2 | 9.4 | |
| GLSCL*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 73.5 | 45.6 | 82.7 | 2 | 9.8 | |
| TMASS†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 73.2 | 45 | 82.7 | 2 | 9.5 | |
| XPool†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 73.1 | 45.7 | 82.3 | 2 | 9.4 | |
| ProST*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 72.9 | 45 | 82.2 | 2 | 8.9 | |
| DREAM2026.06 | 72.9 | 45.7 | 84.2 | 2 | 9.3 | |
| TS2Net*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 72.6 | 44.8 | 81.1 | 2 | 10.3 | |
| CMAYear=20252026.06 | 72.1 | 44.6 | 83.6 | 2 | 9.8 | |
| RAPEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 71.9 | 44 | 82.4 | — | 10.1 | |
| CLIP-VIP*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 71.5 | 43.6 | 82.3 | 2 | 10.8 | |
| TFVLEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 71.5 | 45.5 | 81 | — | 11 | |
| TempMe*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 71.5 | 45 | 80.8 | 2 | 11 | |
| TV-ProxyNet†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 71.1 | 45.4 | 82.4 | 2 | 11 | |
| CLIP4Clip*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 70.9 | 42.5 | 81 | 2 | 12.2 | |
| CLIP4Clip*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 70.9 | 42.9 | 80.6 | 2 | 11.6 | |
| Clip4ClipYear=20222026.06 | 70.5 | 43.1 | 81.2 | 3 | — | |
| Video-ColBERT†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 70.4 | 42.1 | 81.1 | 2 | 10.8 | |
| UMPYear=20242026.06 | 69.9 | 44.8 | 80.6 | — | 11 | |
| Align and TellYear=20232026.06 | 69.3 | 41.6 | 78.6 | 2 | — | |
| TVMMEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 63.8 | 34.8 | 73.7 | 3 | — | |
| TeachTextEnd-to-End Status=Not End-to-End2026.03 | 62.7 | 32.1 | 75 | 3 | — | |
| TT-CE+Year=20212026.06 | 62.7 | 32.1 | 75 | 3 | — | |
| HiTEnd-to-End Status=Not End-to-End2026.03 | 60.3 | 28.8 | 72.3 | 3 | — | |
| T2vladEnd-to-End Status=Not End-to-End2026.03 | 60 | 31.8 | 71.1 | 3 | — | |
| SupportSetEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 58.6 | 28.5 | 71.6 | 3 | — | |
| MMTEnd-to-End Status=Not End-to-End2026.03 | 57.5 | 27 | 69.7 | 3.7 | 21.3 | |
| ESATYear=20232026.06 | 57 | — | 70.3 | 4 | 19.43 | |
| MMTYear=20202026.06 | 56 | 24.4 | 67.8 | 4 | — | |
| Zhu et al.Year=20232026.06 | 53.2 | 25.3 | 63.4 | 5 | — | |
| CEYear=20192026.06 | 50.3 | 20.6 | 64 | 5.3 | 25.1 | |
| CEEnd-to-End Status=Not End-to-End2026.03 | 48.8 | 20.9 | 62.4 | 6 | — |