Text-to-Video Retrieval on ActivityNet
68.2R@1PMRL
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| PMRLsetting=finetuning2025.07 | 68.2 | — | — | — | — | — | — | — | — | |
| VASTsetting=finetuning2025.07 | 68.1 | — | — | — | — | — | — | — | — | |
| GRAMsetting=finetuning2025.07 | 67.6 | — | — | — | — | — | — | — | — | |
| UMT-L#Pairs=25M2023.03 | 66.8 | 0.891 | — | — | — | 94.9 | — | — | — | |
| UMT-L#Pairs=25M2024.03 | 66.8 | — | — | — | — | — | — | — | — | |
| UMT-Lsetting=finetuning, evaluation_frames=122025.07 | 66.8 | — | — | — | — | — | — | — | — | |
| UMT-L + vid-TLDR#Pairs=25M2024.03 | 66.7 | — | — | — | — | — | — | — | — | |
| UMT-L#Pairs=17M2023.03 | 66.6 | 0.886 | — | — | — | 94.7 | — | — | — | |
| vid-TLDRsetting=finetuning, evaluation_frames=122025.07 | 65.2 | — | — | — | — | — | — | — | — | |
| VALOR-Lsetting=finetuning2025.07 | 63.4 | — | — | — | — | — | — | — | — | |
| InternVideo#Pairs=646M2023.03 | 62.2 | 0.859 | — | — | — | 93.2 | — | — | — | |
| InternVideo#Pairs=646M2024.03 | 62.2 | — | — | — | — | — | — | — | — | |
| InternVideo-Lsetting=finetuning, evaluation_frames=122025.07 | 62.2 | — | — | — | — | — | — | — | — | |
| GRAMModality=T-VA, Zero-shot=true2024.12 | 59 | — | — | — | — | — | — | — | — | |
| GRAMModality=T-V, Zero-shot=true2024.12 | 58.9 | — | — | — | — | — | — | — | — | |
| VideoMatch2021.10 | 58.8 | 0.892 | 1 | — | — | — | — | — | — | |
| UMT-B#Pairs=25M2023.03 | 58.3 | 0.839 | — | — | — | 91.5 | — | — | — | |
| UMT-B#Pairs=25M2024.03 | 58.3 | — | — | — | — | — | — | — | — | |
| UMT-L#Pairs=5M2023.03 | 58.1 | 0.855 | — | — | — | 92.9 | — | — | — | |
| UMT-B + vid-TLDR#Pairs=25M2024.03 | 57.8 | — | — | — | — | — | — | — | — | |
| UMT-B#Pairs=17M2023.03 | 56.1 | 0.825 | — | — | — | 91.2 | — | — | — | |
| PMRLZero-shot=true2025.07 | 56 | — | — | — | — | — | — | — | — | |
| VINDLU#Pairs=25M2023.03 | 55 | 0.814 | — | — | — | 89.7 | — | — | — | |
| VINDLU#Pairs=25M2024.03 | 55 | — | — | — | — | — | — | — | — | |
| GRAMZero-shot=true2025.07 | 54.5 | — | — | — | — | — | — | — | — | |
| CLIP-ViP#Pairs=500M2023.03 | 53.4 | 0.814 | — | — | — | 90 | — | — | — | |
| CLIP-ViP#Pairs=500M2024.03 | 53.4 | — | — | — | — | — | — | — | — | |
| TPM-CLBackbone=CLIP-ViT-B/32, DSL post-processing=true, re-training=false2023.09 | 53.4 | 0.807 | 1 | — | — | 89.2 | — | 5.3 | 223.3 | |
| UMT-B#Pairs=5M2023.03 | 52.1 | 0.805 | — | — | — | 89.6 | — | — | — | |
| VASTZero-shot=true2025.07 | 51.7 | — | — | — | — | — | — | — | — | |
| VASTModality=T-VA, Zero-shot=true2024.12 | 51.4 | — | — | — | — | — | — | — | — | |
| CAMoE2021.09 | 51 | 0.777 | — | — | — | — | — | — | — | |
| CAMOEBackbone=CLIP-ViT-B/32, DSL post-processing=true, re-training=false2023.09 | 51 | 0.777 | — | — | — | — | — | — | — | |
| ViCLIPsetting=finetuning2025.07 | 49.8 | — | — | — | — | — | — | — | — | |
| HiTeA#Pairs=17M2023.03 | 49.7 | 0.771 | — | — | — | 86.7 | — | — | — | |
| VideoPrism-bModality=T-V, Zero-shot=true2024.12 | 49.6 | — | — | — | — | — | — | — | — | |
| VideoPrism-bZero-shot=true2025.07 | 49.6 | — | — | — | — | — | — | — | — | |
| Singularity#Pairs=17M2023.03 | 48.9 | 0.77 | — | — | — | 86.3 | — | — | — | |
| Singularity#Pairs=17M2024.03 | 48.9 | — | — | — | — | — | — | — | — | |
| TS2-NetBackbone=CLIP-ViT-B/32, DSL post-processing=true, re-training=true2023.09 | 48.3 | 0.78 | 2 | — | — | 86.8 | — | 7.7 | 213.1 | |
| BaselineBackbone=CLIP-ViT-B/32, DSL post-processing=true, re-training=false2023.09 | 48.3 | 0.763 | 2 | — | — | 86.5 | — | 7.6 | 211.1 | |
| HSE2021.10 | 44.4 | 0.767 | 2 | — | — | — | — | — | — | |
| UMT-LModality=T-V, Zero-shot=true2024.12 | 41.9 | — | — | — | — | — | — | — | — | |
| UMT-LZero-shot=true2025.07 | 41.9 | — | — | — | — | — | — | — | — | |
| LanguageBindModality=T-V, Zero-shot=true2024.12 | 41 | — | — | — | — | — | — | — | — | |
| LanguageBindZero-shot=true2025.07 | 41 | — | — | — | — | — | — | — | — | |
| CLIP4Clip#Pairs=400M2023.03 | 40.5 | 0.724 | — | — | — | 83.4 | — | — | — | |
| CLIP4Clip#Pairs=400M2024.03 | 40.5 | — | — | — | — | — | — | — | — | |
| CLIP4Clipsetting=finetuning2025.07 | 40.5 | — | — | — | — | — | — | — | — | |
| FullParams (M)=119.8 (100%)2022.11 | 36.8 | 66.9 | 3 | 9.3 | — | 80.1 | — | — | — | |
| VoPF+PParams (M)=0.4 (0.328%)2022.11 | 36.1 | 65.5 | 3 | 10.9 | — | 78.5 | — | — | — | |
| VoPF+CParams (M)=14.1 (11.785%)2022.11 | 35.1 | 63.7 | 3 | 11.4 | — | 77.6 | — | — | — | |
| VoPFParams (M)=0.1 (0.103%)2022.11 | 34.6 | 62.6 | 3 | 11.6 | — | 76.4 | — | — | — | |
| VideoCoCaZero-shot=true2025.07 | 34.5 | — | — | — | — | — | — | — | — | |
| PartialParams (M)=7.7 (6.410%)2022.11 | 33.6 | 64 | 3 | 10.6 | — | 77.8 | — | — | — | |
| VoPPParams (M)=0.5 (0.441%)2022.11 | 32.8 | 62.3 | 3 | 12.3 | — | 75.4 | — | — | — | |
| VoPCParams (M)=14.3 (11.898%)2022.11 | 32.6 | 62.5 | 3 | 12 | — | 76.5 | — | — | — | |
| VoPParams (M)=0.1 (0.103%)2022.11 | 32.3 | 61.9 | 3 | 12.4 | — | 75.5 | — | — | — | |
| UMTModality=T-V, Zero-shot=true2024.12 | 31.9 | — | — | — | — | — | — | — | — | |
| UMTZero-shot=true2025.07 | 31.9 | — | — | — | — | — | — | — | — | |
| AdapterFFNParams (M)=2.0 (1.655%)2022.11 | 31.8 | 61 | 3 | 12.8 | — | 75 | — | — | — | |
| AdapterATTNParams (M)=2.0 (1.655%)2022.11 | 31.6 | 60.5 | 3 | 13.1 | — | 74.4 | — | — | — | |
| BiasParams (M)=0.1 (0.104%)2022.11 | 31.3 | 60.3 | 3 | 13.4 | — | 74.2 | — | — | — | |
| InternVideo-LModality=T-V, Zero-shot=true2024.12 | 30.7 | — | — | — | — | — | — | — | — | |
| InternVideo-LZero-shot=true2025.07 | 30.7 | — | — | — | — | — | — | — | — | |
| ProjParams (M)=0.7 (0.547%)2022.11 | 29.8 | 59.1 | 4 | 14.2 | — | 73.3 | — | — | — | |
| HAT-VTRZero-shot adaptation=true, QGS=true2026.02 | 29.18 | 0.5723 | — | — | — | — | — | — | — | |
| MMTPre-training=true2020.07 | 28.7 | 0.614 | 3.3 | 16 | 0.945 | — | — | — | — | |
| TACoLang.=BERT, Video=R-152+S3D-HM2021.08 | 25.8 | 0.563 | 4 | — | — | 93.8 | — | — | — | |
| VideoCoCaModality=T-V, Zero-shot=true2024.12 | 24.5 | — | — | — | — | — | — | — | — | |
| MMTPre-training=false2020.07 | 22.7 | 0.542 | 5 | 20.8 | 0.932 | — | — | — | — | |
| MMTLang.=BERT, Video=Collaborative Experts2021.08 | 22.7 | 0.542 | 5 | — | — | 93.2 | — | — | — | |
| All-in-one#Pairs=138M2023.03 | 22.4 | 0.537 | — | — | — | 67.7 | — | — | — | |
| All-in-one#Pairs=138M2024.03 | 22.4 | — | — | — | — | — | — | — | — | |
| TCRZero-shot adaptation=true, QGS=true2026.02 | 22.11 | 0.483 | — | — | — | — | — | — | — | |
| SARZero-shot adaptation=true, QGS=true2026.02 | 21.8 | 0.4686 | — | — | — | — | — | — | — | |
| TentZero-shot adaptation=true, QGS=true2026.02 | 21.72 | 0.4692 | — | — | — | — | — | — | — | |
| ClipBERT#Pairs=5.4M2023.03 | 21.3 | 0.49 | — | — | — | 63.5 | — | — | — | |
| ClipBERT#Pairs=5.4M2024.03 | 21.3 | — | — | — | — | — | — | — | — | |
| CLIPZero-shot adaptation=true, QGS=true2026.02 | 21.17 | 0.4635 | — | — | — | — | — | — | — | |
| HSE2020.07 | 20.5 | 0.493 | — | — | — | — | — | — | — | |
| EATAZero-shot adaptation=true, QGS=true2026.02 | 18.32 | 0.4257 | — | — | — | — | — | — | — | |
| FSE2021.10 | 18.2 | 0.448 | 7 | — | — | — | — | — | — | |
| FSE2020.07 | 18.2 | 0.448 | 7 | — | 0.891 | — | — | — | — | |
| CE2020.07 | 18.2 | 0.477 | 6 | 23.1 | 0.914 | — | — | — | — | |
| FSELang.=GRU, Video=C3D+TSN-Inception2021.08 | 18.2 | 0.448 | 7 | — | — | 89.1 | — | — | — | |
| CELang.=GPT, Video=Collaborative Experts2021.08 | 18.2 | 0.477 | 6 | — | — | 91.4 | — | — | — | |
| READZero-shot adaptation=true, QGS=true2026.02 | 15.44 | 0.3647 | — | — | — | — | — | — | — | |
| ViCLIPModality=T-V, Zero-shot=true2024.12 | 15.1 | — | — | — | — | — | — | — | — | |
| ViCLIPZero-shot=true2025.07 | 15.1 | — | — | — | — | — | — | — | — | |
| Dense full2021.10 | 14 | 0.32 | 34 | — | — | — | — | — | — | |
| DenseCapLang.=LSTM, Video=C3D2021.08 | 14 | 0.32 | 34 | — | — | 65 | — | — | — | |
| No Context2021.10 | 5 | 0.14 | 78 | — | — | — | — | — | — | |
| InternVideo2s2-6BFinetuning=true2024.03 | 0.741 | — | — | — | — | — | — | — | — | |
| VASTSample=442M, Modality=Audio/Subtitle2023.05 | 0.705 | 0.909 | — | — | — | 0.955 | — | — | — | |
| VAST#Pairs=152M2026.03 | 0.705 | — | — | — | — | — | — | — | — | |
| GRAM ModelModality=T-VA, Protocol=Finetuning2024.12 | 0.699 | — | — | — | — | 0.961 | — | — | — | |
| GRAM ModelModality=T-VA, Finetuning=true2024.12 | 0.699 | — | — | — | — | — | — | — | — | |
| VASTModality=T-VA, Protocol=Finetuning2024.12 | 0.688 | — | — | — | — | 0.955 | — | — | — | |
| VASTModality=T-VA, Finetuning=true2024.12 | 0.688 | — | — | — | — | — | — | — | — |