Text-to-Video Retrieval on MSRVTT (R@1, R@5, R@10, MdR, MnR, Rsum)
51Recall@1EagleNet
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| EagleNetEnd-to-End Status=End-to-End, Backbone=ViT-B/162026.03 | 51 | 76.2 | 85.6 | 1 | 12.8 | 425.7 | |
| Video-ColBERT†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 50 | 76.3 | 84.3 | 1.5 | 13.1 | 417.8 | |
| GLSCL*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 49.9 | 76.3 | 84.1 | 2 | 11.9 | 419 | |
| TempMe*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 49.9 | 75.3 | 84.8 | 2 | 12.1 | 417.8 | |
| TV-ProxyNet†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 49.5 | 74.9 | 83.4 | 2 | 12.6 | 414.6 | |
| XPool†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 49.2 | 73.9 | 82.6 | 2 | 13.6 | 411.5 | |
| TS2Net*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 48.8 | 75 | 85 | 2 | 13 | 415.9 | |
| TFVLEnd-to-End Status=End-to-End, Backbone=ViT-B/162026.03 | 48.7 | 73.5 | 83 | — | 13.4 | 411.5 | |
| UCoFiA*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 48.6 | 75.5 | 83.8 | 2 | 11.7 | 416.8 | |
| CLIP-VIP*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 48.5 | 75.5 | 83.9 | 2 | 12.5 | 415.3 | |
| TMASS†*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 48.5 | 74.1 | 83.4 | 2 | 14.4 | 414.2 | |
| ProST*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 48 | 74.7 | 84 | 2 | 12.7 | 414.7 | |
| GLSCL*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 47.2 | 73.5 | 82.7 | 2 | 13.1 | 405.2 | |
| EagleNetEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 47.2 | 74.8 | 84 | 2 | 14.4 | 408.8 | |
| TV-ProxyNet†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 46.7 | 73 | 81.8 | 2 | 14.5 | 400.4 | |
| RAPEnd-to-End Status=End-to-End, Backbone=ViT-B/162026.03 | 46.5 | 73.9 | 82 | — | 12.1 | 408.9 | |
| ProST*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 46.3 | 73.6 | 82.4 | 2 | 12.9 | 402.4 | |
| CLIP-VIP*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 46 | 72.6 | 82.2 | 2 | 14.4 | 398.2 | |
| Video-ColBERT†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 45.9 | 71.6 | 82.7 | 2 | 13.7 | 393.8 | |
| XPool†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 45.8 | 73.8 | 82 | 2 | 14 | 402.7 | |
| TS2Net*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 45.4 | 72.9 | 82.9 | 2 | 14.2 | 399.7 | |
| SHE-NetEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 45.3 | 74.9 | 84.2 | 2 | 12.5 | 407.7 | |
| TFVLEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 45.2 | 70.4 | 80.1 | — | 15.2 | 393.7 | |
| CLIP4Clip*End-to-End Status=End-to-End, Backbone=ViT-B/16, Fine-tuning=true2026.03 | 45.2 | 72.2 | 81.4 | 2 | 15 | 393.2 | |
| RAPEnd-to-End Status=End-to-End, Backbone=ViT-B/322026.03 | 44.8 | 71.4 | 81.5 | — | 14.4 | 396 | |
| TMASS†*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 43.4 | 73.2 | 82.1 | 2 | 14.3 | 399.6 | |
| TempMe*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 43.3 | 71.2 | 81.2 | 2 | 15.3 | 393 | |
| CLIP4Clip*End-to-End Status=End-to-End, Backbone=ViT-B/32, Fine-tuning=true2026.03 | 43.2 | 71.4 | 80.6 | 2 | 16.5 | 389.6 | |
| BridgeFormerEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 37.6 | 64.8 | 75.1 | 3 | — | — | |
| TVMMEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 36.2 | 64.2 | 75.7 | 3 | — | 348.4 | |
| FrozenEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 31 | 59.5 | 70.5 | 3 | — | — | |
| SupportSetEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 30.1 | 58.5 | 69.3 | 3 | — | 316.6 | |
| TeachTextEnd-to-End Status=Not End-to-End2026.03 | 29.6 | 61.6 | 74.2 | 3 | — | 335.2 | |
| T2vladEnd-to-End Status=Not End-to-End2026.03 | 29.5 | 59 | 70.1 | 4 | — | 321.5 | |
| HiTEnd-to-End Status=Not End-to-End2026.03 | 27.7 | 59.2 | 72 | 2.9 | — | 320.3 | |
| MMTEnd-to-End Status=Not End-to-End2026.03 | 26.6 | 57.1 | 69.6 | 4 | 24 | 307.5 | |
| ClipBERTEnd-to-End Status=End-to-End, Backbone=non-CLIP2026.03 | 22 | 46.8 | 59.9 | 6 | — | — | |
| CEEnd-to-End Status=Not End-to-End2026.03 | 20.6 | 50.3 | 64 | 5.3 | — | 267 |