Video Retrieval on MSR-VTT
57.7R@1Absolute SOTA
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Absolute SOTAReference=[79]2023.05 | 57.7 | — | — | — | — | |
| AGCTok=322026.02 | 56.9 | 87 | 71.5 | — | — | |
| AGCTok=1282026.02 | 56.4 | 87.6 | 71.6 | — | — | |
| Baseline 3B (Ours)Tok=13182026.02 | 55.7 | 88.3 | 71.9 | — | — | |
| MemTokTok=1282026.02 | 54.9 | 86.7 | 70.5 | — | — | |
| H-PoolTok=1282026.02 | 54.4 | 87.2 | 70.9 | — | — | |
| MemTokTok=322026.02 | 54.2 | 86.4 | 69.9 | — | — | |
| H-PoolTok=322026.02 | 54.1 | 87.3 | 70.4 | — | — | |
| AGCTok=52026.02 | 53.9 | 85.8 | 69.2 | — | — | |
| SeqResizeTok=52026.02 | 53.4 | 86.7 | 69.5 | — | — | |
| SeqResizeTok=322026.02 | 53.3 | 86.9 | 69.9 | — | — | |
| MemTokTok=52026.02 | 52.7 | 87.3 | 69.3 | — | — | |
| H-PoolTok=52026.02 | 52.6 | 86.1 | 68.9 | — | — | |
| SeqResizeTok=1282026.02 | 52.6 | 87.6 | 69.7 | — | — | |
| OmniEmbed-7BTok=12026.02 | 51.5 | 83.2 | 67.1 | — | — | |
| Video-ColBERTTok=262026.02 | 51.5 | 85.5 | 67.7 | — | — | |
| LanguageBind2026.05 | 44.8 | — | — | — | — | |
| ViProBase Model=DRL2025.08 | 42.35 | — | — | 78.39 | 60.37 | |
| ColQwen-Omni 3BTok=17022026.02 | 40.8 | 73.8 | 56.3 | — | — | |
| SGABase Model=DRL2025.08 | 39.41 | — | — | 76.46 | 57.97 | |
| Co-AttackBase Model=DRL2025.08 | 38.61 | — | — | 76.09 | 57.35 | |
| CodeBind-IB2026.05 | 37.8 | — | — | — | — | |
| UNIALIGN2026.05 | 37 | — | — | — | — | |
| IMAGEBINDZero-shot=true, Text Prompting=true2023.05 | 36.1 | — | — | — | — | |
| ViProBase Model=Sing2025.08 | 8.02 | — | — | 28.31 | 18.17 | |
| No AttackBase Model=Sing2025.08 | 4.87 | — | — | 22.94 | 13.91 | |
| Co-AttackBase Model=Sing2025.08 | 4.66 | — | — | 20.86 | 12.76 | |
| No AttackBase Model=DRL2025.08 | 4.29 | — | — | 15.76 | 10.03 | |
| No AttackBase Model=C4V2025.08 | 3.76 | — | — | 13.42 | 8.59 | |
| ViProBase Model=C4V2025.08 | 1.99 | — | — | 10.53 | 6.26 | |
| SGABase Model=C4V2025.08 | 1.28 | — | — | 7.69 | 4.49 | |
| SGABase Model=Sing2025.08 | 0.98 | — | — | 4.05 | 2.52 | |
| Co-AttackBase Model=C4V2025.08 | 0.93 | — | — | 5.07 | 3 | |
| RandomZero-shot=true2023.05 | 0.1 | — | — | — | — |