Text-to-Video Retrieval on VATEX
95.1R@1PEAV-L
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| PEAV-LV-Enc Params=0.5B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.12 | 95.1 | — | — | — | — | — | |
| PEAV-LV-Enc Params=0.5B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 94.9 | — | — | — | — | — | |
| PEAV-BV-Enc Params=0.4B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.12 | 94.9 | — | — | — | — | — | |
| PEAV-SV-Enc Params=0.3B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.12 | 94.5 | — | — | — | — | — | |
| PEAV-BV-Enc Params=0.4B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 94.3 | — | — | — | — | — | |
| PEAV-SV-Enc Params=0.3B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 94.2 | — | — | — | — | — | |
| GRAM ModelModality=T-VAS, Protocol=Finetuning2024.12 | 87.7 | — | 100 | — | — | — | |
| GRAM ModelModality=T-VAS, Finetuning=true2024.12 | 87.7 | — | — | — | — | — | |
| VASTModality=T-VAS, Protocol=Finetuning2024.12 | 87.5 | — | 99.5 | — | — | — | |
| VASTModality=T-VAS, Finetuning=true2024.12 | 87.5 | — | — | — | — | — | |
| GRAM ModelModality=T-VA, Protocol=Finetuning2024.12 | 87 | — | 99.5 | — | — | — | |
| GRAM ModelModality=T-VA, Finetuning=true2024.12 | 87 | — | — | — | — | — | |
| VASTModality=T-VA, Protocol=Finetuning2024.12 | 86.9 | — | 99.1 | — | — | — | |
| VASTModality=T-VA, Finetuning=true2024.12 | 86.9 | — | — | — | — | — | |
| GRAM ModelModality=T-V, Protocol=Finetuning2024.12 | 84.4 | — | 99.8 | — | — | — | |
| GRAM ModelModality=T-V, Finetuning=true2024.12 | 84.4 | — | — | — | — | — | |
| PMRLsetting=finetuning2025.07 | 84.1 | — | — | — | — | — | |
| GRAMModality=T-VA, Zero-shot=true2024.12 | 83.9 | — | — | — | — | — | |
| GRAM ModelModality=T-VA, Zero-shot=true2024.12 | 83.9 | — | 98.6 | — | — | — | |
| PE-GV-Enc Params=1.9B, Video Data=22M, Resolution=448, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 83.6 | — | — | — | — | — | |
| GRAMModality=T-VAS, Zero-shot=true2024.12 | 83.5 | — | — | — | — | — | |
| GRAM ModelModality=T-VAS, Zero-shot=true2024.12 | 83.5 | — | 98.8 | — | — | — | |
| VASTsetting=finetuning2025.07 | 83.1 | — | — | — | — | — | |
| VASTModality=Audio/Subtitle2023.05 | 83 | 98.2 | 99.2 | — | — | — | |
| LanguageBindV-Enc Params=0.3B, Video Data=10M, Resolution=224, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 82.9 | — | — | — | — | — | |
| GRAMsetting=finetuning2025.07 | 82.5 | — | — | — | — | — | |
| VASTModality=T-VAS, Zero-shot=true2024.12 | 82.1 | — | — | — | — | — | |
| VASTModality=T-VAS, Zero-shot=true2024.12 | 82.1 | — | 96.8 | — | — | — | |
| GRAMModality=T-V, Zero-shot=true2024.12 | 81.1 | — | — | — | — | — | |
| GRAM ModelModality=T-V, Zero-shot=true2024.12 | 81.1 | — | 99.5 | — | — | — | |
| PMRLZero-shot=true2025.07 | 80.5 | — | — | — | — | — | |
| VASTModality=T-VA, Zero-shot=true2024.12 | 80 | — | — | — | — | — | |
| VASTModality=T-VA, Zero-shot=true2024.12 | 80 | — | 95.7 | — | — | — | |
| PE-LV-Enc Params=0.3B, Video Data=22M, Resolution=336, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 79 | — | — | — | — | — | |
| COSAExample=415M2023.06 | 78.6 | 97.1 | 98.7 | — | — | — | |
| GRAMZero-shot=true2025.07 | 77.5 | — | — | — | — | — | |
| VALORExample=433.5M2023.06 | 76.9 | 96.7 | 98.6 | — | — | — | |
| VALOR-LModality=Audio/Subtitle2023.05 | 76.9 | 96.7 | 98.6 | — | — | — | |
| VALOR-LModality=T-VAS, Finetuning=true2024.12 | 76.9 | — | — | — | — | — | |
| VALOR-Lsetting=finetuning2025.07 | 76.9 | — | — | — | — | — | |
| VASTZero-shot=true2025.07 | 75.9 | — | — | — | — | — | |
| VidLABackbone=CLIP-ViT-B/16, dual-softmax inference=true2024.03 | 75.8 | 88.3 | 89.3 | — | — | — | |
| InternVideo2s2-6BFinetuning=true2024.03 | 75.5 | — | — | — | — | — | |
| COSA-LExample=417M2023.06 | 75.4 | 95.9 | 98.2 | — | — | — | |
| VidLABackbone=CLIP-ViT-B/32, dual-softmax inference=true2024.03 | 73.7 | 87.6 | 89.1 | — | — | — | |
| UMT-LFinetuning=true2024.03 | 72 | — | — | — | — | — | |
| UMT-LModality=T-V, Protocol=Finetuning, Number of frames=122024.12 | 72 | — | — | — | — | — | |
| UMT-LModality=T-V, Finetuning=true, Evaluation frames=122024.12 | 72 | — | — | — | — | — | |
| UMT-Lsetting=finetuning, evaluation_frames=122025.07 | 72 | — | — | — | — | — | |
| InternVideo2-6BV-T Pairs=100M2026.02 | 71.5 | — | — | — | — | — | |
| InternVideo(ViT-L)Backbone=ViT-L, dual-softmax inference=true2024.03 | 71.1 | — | — | — | — | — | |
| InternVideo-LModality=T-V, Protocol=Finetuning, Number of frames=122024.12 | 71.1 | — | — | — | — | — | |
| InternVideo-LModality=T-V, Finetuning=true, Evaluation frames=122024.12 | 71.1 | — | — | — | — | — | |
| InternVideo-Lsetting=finetuning, evaluation_frames=122025.07 | 71.1 | — | — | — | — | — | |
| InternVideo2V-Enc Params=1.0B, Video Data=102M, Resolution=224, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 70.4 | — | — | — | — | — | |
| InternVideo2-1BSetting=ZS2026.05 | 70.4 | — | — | — | — | — | |
| VidVecV-T Pairs=Text-Only2026.02 | 70 | — | — | — | — | — | |
| ImageBindV-Enc Params=0.6B, Video Data=3M, Resolution=224, Frame/ fps=16, Zero-shot Protocol=true2025.12 | 69.8 | — | — | — | — | — | |
| Gemini Embedding 2Setting=ZS2026.05 | 69.4 | — | — | — | — | — | |
| VidLABackbone=CLIP-ViT-B/162024.03 | 69.2 | 87.1 | 88.9 | — | — | — | |
| NarVid2025.03 | 68.4 | 94 | 97.1 | — | — | — | |
| COSA-BExample=17M2023.06 | 67.7 | 93.4 | 96.9 | — | — | — | |
| SigLIP2-g-optV-Enc Params=1.1B, Resolution=384, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 67.5 | — | — | — | — | — | |
| Cap4VideoBackbone=CLIP-ViT-B/162024.03 | 66.6 | 93.1 | 97 | — | — | — | |
| Cap4VideoVenue=CVPR'232025.03 | 66.6 | 93.1 | 97 | — | — | — | |
| VidLABackbone=CLIP-ViT-B/322024.03 | 66.5 | 86.2 | 88.4 | — | — | — | |
| DCRExample=400M2023.06 | 65.7 | 92.6 | 96.7 | — | — | — | |
| DCRModality=Vision-only2023.05 | 65.7 | 92.6 | 96.7 | — | — | — | |
| DRLBackbone=CLIP-ViT-B/162024.03 | 65.7 | 92.6 | 96.7 | — | — | — | |
| T-MASSModality=T-VA, Protocol=Finetuning2024.12 | 65.6 | — | 97.2 | — | — | — | |
| T-MASSModality=T-VA, Finetuning=true2024.12 | 65.6 | — | — | — | — | — | |
| T-MASSVenue=CVPR'242025.03 | 65.6 | 93.9 | 97.2 | — | — | — | |
| T-MASSsetting=finetuning2025.07 | 65.6 | — | — | — | — | — | |
| COSA-BExample=5M2023.06 | 65.3 | 92 | 96 | — | — | — | |
| UATVRBackbone=CLIP-ViT-B/162024.03 | 64.5 | 92.6 | 96.8 | — | — | — | |
| UATVRVenue=ICCV'232025.03 | 64.5 | 92.6 | 96.8 | — | — | — | |
| SigLIP2-L/16V-Enc Params=0.3B, Resolution=384, Frame/ fps=8, Zero-shot Protocol=true2025.12 | 64.1 | — | — | — | — | — | |
| TV-ProxyNetBackbone=ViT-B/16, Fine-tuning=true2026.03 | 64 | 91.7 | 95.9 | — | — | — | |
| CLIP-VIPBackbone=ViT-B/16, Fine-tuning=true2026.03 | 63.8 | 91.6 | 96 | — | — | — | |
| EagleNetBackbone=ViT-B/162026.03 | 63.6 | 91.5 | 95.7 | — | — | — | |
| DRLBackbone=CLIP-ViT-B/322024.03 | 63.5 | 91.7 | 96.5 | — | — | — | |
| CLIP4ClipModality=T-V, Protocol=Finetuning2024.12 | 63 | — | — | — | — | — | |
| CLIP4ClipModality=T-V, Finetuning=true2024.12 | 63 | — | — | — | — | — | |
| VideoPrism-bModality=T-V, Zero-shot=true2024.12 | 62.5 | — | — | — | — | — | |
| VideoPrism-bModality=T-V, Zero-shot=true2024.12 | 62.5 | — | — | — | — | — | |
| VideoPrism-gV-T Pairs=600M2026.02 | 62.5 | — | — | — | — | — | |
| VideoPrism-bZero-shot=true2025.07 | 62.5 | — | — | — | — | — | |
| TS2NetBackbone=ViT-B/16, Fine-tuning=true2026.03 | 62.1 | 91.2 | 95.8 | — | — | — | |
| XPoolBackbone=ViT-B/16, Fine-tuning=true2026.03 | 62 | 91 | 95.6 | — | — | — | |
| ProSTBackbone=ViT-B/16, Fine-tuning=true2026.03 | 62 | 91.3 | 95.8 | — | — | — | |
| UATVRBackbone=CLIP-ViT-B/322024.03 | 61.3 | 91 | 95.6 | — | — | — | |
| Video-ColBERTBackbone=ViT-B/16, Fine-tuning=true2026.03 | 61.1 | 90.2 | 95.3 | — | — | — | |
| TEFALModality=T-VA, Protocol=Finetuning2024.12 | 61 | — | 95.3 | — | — | — | |
| TEFALVenue=ICCV'232025.03 | 61 | 90.4 | 95.3 | — | — | — | |
| TEFALsetting=finetuning2025.07 | 61 | — | — | — | — | — | |
| CLIP4ClipBackbone=ViT-B/16, Fine-tuning=true2026.03 | 60.9 | 90.6 | 95.4 | — | — | — | |
| TMASSBackbone=ViT-B/16, Fine-tuning=true2026.03 | 60.8 | 90.5 | 95.3 | — | — | — | |
| CLIP-VIPBackbone=ViT-B/32, Fine-tuning=true2026.03 | 59.3 | 89 | 94.3 | — | — | — | |
| EagleNetBackbone=ViT-B/322026.03 | 59 | 89.2 | 94.1 | — | — | — | |
| TV-ProxyNetBackbone=ViT-B/32, Fine-tuning=true2026.03 | 58.9 | 89.1 | 94.4 | — | — | — |