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/HiTeA

HiTeA

Reported on 51 benchmarks across 6 tasks · 1 paper · 14 SOTA

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

Computer Vision44 results

VideoonSSv2-template retrieval
text-to-video R@1· uses extra data· 2022-12-30
85.6
best: 90.8 (UMT-L (ViT-L/16))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonSSv2-label retrieval
text-to-video R@1· uses extra data· 2022-12-30
55.2
best: 73.3 (UMT-L (ViT-L/16))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonSSv2-label retrieval
text-to-video R@5· uses extra data· 2022-12-30
89.1
best: 93.3 (vid-TLDR (UMT-L))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSVD
BLEU-4· uses extra data· 2022-12-30
71
best: 80.7 (VALOR)
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSVD
CIDEr· uses extra data· 2022-12-30
146.9
best: 195.6 (MaMMUT)
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSVD
METEOR· uses extra data· 2022-12-30
45.3
best: 51.2 (VLAB)
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSVD
ROUGE-L· uses extra data· 2022-12-30
81.4
best: 87.9 (VLAB)
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonSSv2-template retrieval
text-to-video R@1· uses extra data· 2022-12-30
85.6
best: 90.8 (UMT-L (ViT-L/16))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonSSv2-label retrieval
text-to-video R@1· uses extra data· 2022-12-30
55.2
best: 73.3 (UMT-L (ViT-L/16))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonSSv2-label retrieval
text-to-video R@5· uses extra data· 2022-12-30
89.1
best: 93.3 (vid-TLDR (UMT-L))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonMSR-VTT-1kA
text-to-video R@1· 2022-12-30
46.8
best: 62.9 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonMSR-VTT-1kA
text-to-video R@10· 2022-12-30
81.9
best: 90.8 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonMSR-VTT-1kA
text-to-video R@5· 2022-12-30
71.2
best: 84.5 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonSSv2-template retrieval
text-to-video R@10· uses extra data· 2022-12-30
100
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonSSv2-template retrieval
text-to-video R@5· uses extra data· 2022-12-30
100
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonActivityNet
text-to-video R@1· uses extra data· 2022-12-30
49.7
best: 74.1 (InternVideo2-6B)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonActivityNet
text-to-video R@10· uses extra data· 2022-12-30
86.7
best: 96.1 (GRAM)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonActivityNet
text-to-video R@5· uses extra data· 2022-12-30
77.1
best: 90.9 (VAST)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonSSv2-label retrieval
text-to-video R@10· uses extra data· 2022-12-30
81.4
best: 96.6 (UMT-L (ViT-L/16))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonDiDeMo
text-to-video R@1· uses extra data· 2022-12-30
56.5
best: 74.2 (InternVideo2-6B)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonDiDeMo
text-to-video R@10· uses extra data· 2022-12-30
89.7
best: 94.2 (vid-TLDR (UMT-L))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonDiDeMo
text-to-video R@5· uses extra data· 2022-12-30
81.7
best: 91.2 (vid-TLDR (UMT-L))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonLSMDC
text-to-video R@1· uses extra data· 2022-12-30
28.7
best: 46.4 (InternVideo2-6B)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonLSMDC
text-to-video R@10· uses extra data· 2022-12-30
59
best: 92.8 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
VideoonLSMDC
text-to-video R@5· uses extra data· 2022-12-30
50.3
best: 80.1 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSR-VTT
BLEU-4· uses extra data· 2022-12-30
49.2
best: 57.8 (mPLUG-2)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSR-VTT
CIDEr· uses extra data· 2022-12-30
65.1
best: 80 (mPLUG-2)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSR-VTT
METEOR· uses extra data· 2022-12-30
30.7
best: 38.7 (MV-GPT)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video CaptioningonMSR-VTT
ROUGE-L· uses extra data· 2022-12-30
65
best: 70.1 (mPLUG-2)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonMSR-VTT-1kA
text-to-video R@1· 2022-12-30
46.8
best: 62.9 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonMSR-VTT-1kA
text-to-video R@10· 2022-12-30
81.9
best: 90.8 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonMSR-VTT-1kA
text-to-video R@5· 2022-12-30
71.2
best: 84.5 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonSSv2-template retrieval
text-to-video R@10· uses extra data· 2022-12-30
100
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonSSv2-template retrieval
text-to-video R@5· uses extra data· 2022-12-30
100
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonActivityNet
text-to-video R@1· uses extra data· 2022-12-30
49.7
best: 74.1 (InternVideo2-6B)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonActivityNet
text-to-video R@10· uses extra data· 2022-12-30
86.7
best: 96.1 (GRAM)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonActivityNet
text-to-video R@5· uses extra data· 2022-12-30
77.1
best: 90.9 (VAST)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonSSv2-label retrieval
text-to-video R@10· uses extra data· 2022-12-30
81.4
best: 96.6 (UMT-L (ViT-L/16))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonDiDeMo
text-to-video R@1· uses extra data· 2022-12-30
56.5
best: 74.2 (InternVideo2-6B)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonDiDeMo
text-to-video R@10· uses extra data· 2022-12-30
89.7
best: 94.2 (vid-TLDR (UMT-L))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonDiDeMo
text-to-video R@5· uses extra data· 2022-12-30
81.7
best: 91.2 (vid-TLDR (UMT-L))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonLSMDC
text-to-video R@1· uses extra data· 2022-12-30
28.7
best: 46.4 (InternVideo2-6B)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonLSMDC
text-to-video R@10· uses extra data· 2022-12-30
59
best: 92.8 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video RetrievalonLSMDC
text-to-video R@5· uses extra data· 2022-12-30
50.3
best: 80.1 (HunYuan_tvr (huge))
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546

Natural Language Processing3 results

Visual Question Answering (VQA)onTGIF-QA
Accuracy· 2022-12-30
0.732
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Visual Question Answering (VQA)onMSRVTT-QA
Accuracy· uses extra data· 2022-12-30
0.459
best: 0.496 (VLAB)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Visual Question Answering (VQA)onMSVD-QA
Accuracy· uses extra data· 2022-12-30
0.556
best: 0.61 (VLAB)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546

Methodology2 results

Zero-Shot LearningonMSRVTT-QA
Accuracy· uses extra data· 2022-12-30
21.7
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Zero-Shot LearningonMSVD-QA
Accuracy· uses extra data· 2022-12-30
37.4
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546

Reasoning2 results

Video Question AnsweringonNExT-QA
Accuracy· uses extra data· 2022-12-30
63.1
best: 85.5 (LinVT-Qwen2-VL (7B))
SOTA
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546
Video Question AnsweringonMSRVTT-MC
Accuracy· 2022-12-30
97.4
best: 97.6 (VIOLETv2)
HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training arXiv:2212.14546