Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/VindLU

VindLU

Reported on 32 benchmarks across 3 tasks · 1 paper · 16 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision28 results

VideoonSSv2-template retrieval
text-to-video R@1· uses extra data· 2022-12-09
83.3
best: 90.8 (UMT-L (ViT-L/16))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonSSv2-template retrieval
text-to-video R@10· uses extra data· 2022-12-09
100
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonSSv2-template retrieval
text-to-video R@5· uses extra data· 2022-12-09
100
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonSSv2-label retrieval
text-to-video R@1· uses extra data· 2022-12-09
53.1
best: 73.3 (UMT-L (ViT-L/16))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonSSv2-label retrieval
text-to-video R@5· uses extra data· 2022-12-09
81.8
best: 93.3 (vid-TLDR (UMT-L))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonDiDeMo
text-to-video R@1· uses extra data· 2022-12-09
61.2
best: 74.2 (InternVideo2-6B)
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonDiDeMo
text-to-video R@10· uses extra data· 2022-12-09
91
best: 94.2 (vid-TLDR (UMT-L))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonDiDeMo
text-to-video R@5· uses extra data· 2022-12-09
85.8
best: 91.2 (vid-TLDR (UMT-L))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonSSv2-template retrieval
text-to-video R@1· uses extra data· 2022-12-09
83.3
best: 90.8 (UMT-L (ViT-L/16))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonSSv2-template retrieval
text-to-video R@10· uses extra data· 2022-12-09
100
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonSSv2-template retrieval
text-to-video R@5· uses extra data· 2022-12-09
100
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonSSv2-label retrieval
text-to-video R@1· uses extra data· 2022-12-09
53.1
best: 73.3 (UMT-L (ViT-L/16))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonSSv2-label retrieval
text-to-video R@5· uses extra data· 2022-12-09
81.8
best: 93.3 (vid-TLDR (UMT-L))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonDiDeMo
text-to-video R@1· uses extra data· 2022-12-09
61.2
best: 74.2 (InternVideo2-6B)
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonDiDeMo
text-to-video R@10· uses extra data· 2022-12-09
91
best: 94.2 (vid-TLDR (UMT-L))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonDiDeMo
text-to-video R@5· uses extra data· 2022-12-09
85.8
best: 91.2 (vid-TLDR (UMT-L))
SOTA
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonMSR-VTT-1kA
text-to-video R@1· uses extra data· 2022-12-09
46.5
best: 62.9 (HunYuan_tvr (huge))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonMSR-VTT-1kA
text-to-video R@10· uses extra data· 2022-12-09
80.4
best: 90.8 (HunYuan_tvr (huge))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonMSR-VTT-1kA
text-to-video R@5· uses extra data· 2022-12-09
71.5
best: 84.5 (HunYuan_tvr (huge))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonActivityNet
text-to-video R@1· uses extra data· 2022-12-09
55
best: 74.1 (InternVideo2-6B)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonActivityNet
text-to-video R@10· uses extra data· 2022-12-09
89.7
best: 96.1 (GRAM)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
VideoonActivityNet
text-to-video R@5· uses extra data· 2022-12-09
81.4
best: 90.9 (VAST)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonMSR-VTT-1kA
text-to-video R@1· uses extra data· 2022-12-09
46.5
best: 62.9 (HunYuan_tvr (huge))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonMSR-VTT-1kA
text-to-video R@10· uses extra data· 2022-12-09
80.4
best: 90.8 (HunYuan_tvr (huge))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonMSR-VTT-1kA
text-to-video R@5· uses extra data· 2022-12-09
71.5
best: 84.5 (HunYuan_tvr (huge))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonActivityNet
text-to-video R@1· uses extra data· 2022-12-09
55
best: 74.1 (InternVideo2-6B)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonActivityNet
text-to-video R@10· uses extra data· 2022-12-09
89.7
best: 96.1 (GRAM)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video RetrievalonActivityNet
text-to-video R@5· uses extra data· 2022-12-09
81.4
best: 90.9 (VAST)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051

Reasoning4 results

Video Question AnsweringonTVQA
Accuracy· uses extra data· 2022-12-09
79
best: 82.2 (LLaMA-VQA)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video Question AnsweringonActivityNet-QA
Accuracy· uses extra data· 2022-12-09
44.7
best: 61.6 (Tarsier (34B))
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video Question AnsweringonMSRVTT-QA
Accuracy· uses extra data· 2022-12-09
44.6
best: 72.4 (Flash-VStream)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051
Video Question AnsweringonMSRVTT-MC
Accuracy· 2022-12-09
95.5
best: 97.6 (VIOLETv2)
VindLU: A Recipe for Effective Video-and-Language Pretraining arXiv:2212.05051