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/Florence

Florence

Reported on 41 benchmarks across 11 tasks · 1 paper · 20 SOTA

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

Miscellaneous16 results

Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· uses extra data· 2021-11-22
81.8
best: 84.8 (BEiT-3)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2021-11-22
81.8
best: 84.8 (BEiT-3)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· uses extra data· 2021-11-22
95.2
best: 96.5 (X2-VLM (large))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· uses extra data· 2021-11-22
63.2
best: 68 (VAST)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· uses extra data· 2021-11-22
85.7
best: 92.8 (BEiT-3)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@1· 2021-11-22
90.9
best: 98.8 (X2-VLM (large))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@5· 2021-11-22
99.1
best: 100 (X2-VLM (large))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@1· 2021-11-22
76.7
best: 93.3 (ERNIE-ViL 2.0)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@5· 2021-11-22
93.6
best: 99.5 (M2-Encoder)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· 2021-11-22
64.7
best: 84.8 (BEiT-3)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· 2021-11-22
85.9
best: 96.5 (X2-VLM (large))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· 2021-11-22
47.2
best: 68 (VAST)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· 2021-11-22
71.4
best: 92.8 (BEiT-3)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2021-11-22
95.2
best: 96.5 (X2-VLM (large))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2021-11-22
63.2
best: 68 (VAST)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2021-11-22
85.7
best: 92.8 (BEiT-3)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432

Computer Vision13 results

Action Recognition In VideosonKinetics-600
Top-1 Accuracy· 2021-11-22
87.8
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Action Recognition In VideosonKinetics-600
Top-5 Accuracy· 2021-11-22
97.8
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Action Recognition In VideosonKinetics-400
Top-1 Accuracy· 2021-11-22
86.5
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Action Recognition In VideosonKinetics-400
Top-5 Accuracy· 2021-11-22
97.3
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Zero-Shot Video RetrievalonMSR-VTT
text-to-video R@1· 2021-11-22
37.6
best: 55.9 (InternVideo2-6B)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Zero-Shot Video RetrievalonMSR-VTT
text-to-video R@10· 2021-11-22
72.6
best: 85.1 (InternVideo2-6B)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Zero-Shot Video RetrievalonMSR-VTT
text-to-video R@5· 2021-11-22
63.8
best: 78.3 (InternVideo2-6B)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
VideoonMSR-VTT-1kA
text-to-video R@1· uses extra data· 2021-11-22
37.6
best: 62.9 (HunYuan_tvr (huge))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
VideoonMSR-VTT-1kA
text-to-video R@10· uses extra data· 2021-11-22
72.6
best: 90.8 (HunYuan_tvr (huge))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
VideoonMSR-VTT-1kA
text-to-video R@5· uses extra data· 2021-11-22
63.8
best: 84.5 (HunYuan_tvr (huge))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Video RetrievalonMSR-VTT-1kA
text-to-video R@1· uses extra data· 2021-11-22
37.6
best: 62.9 (HunYuan_tvr (huge))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Video RetrievalonMSR-VTT-1kA
text-to-video R@10· uses extra data· 2021-11-22
72.6
best: 90.8 (HunYuan_tvr (huge))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Video RetrievalonMSR-VTT-1kA
text-to-video R@5· uses extra data· 2021-11-22
63.8
best: 84.5 (HunYuan_tvr (huge))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432

Natural Language Processing8 results

Cross-Modal RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2021-11-22
81.8
best: 84.8 (BEiT-3)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Visual Question AnsweringonVQA v2 test-dev
Accuracy· 2021-11-22
80.16
best: 82.3 (BLIP-2 ViT-G OPT 6.7B (fine-tuned))
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Visual Question AnsweringonVQA v2 test-std
overall· 2021-11-22
80.36
best: 81.98 (OFA)
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Visual Question Answering (VQA)onVQA v2 test-dev
Accuracy· 2021-11-22
80.16
best: 84.3 (PaLI)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Visual Question Answering (VQA)onVQA v2 test-std
overall· 2021-11-22
80.36
best: 84.03 (BEiT-3)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2021-11-22
95.2
best: 96.5 (X2-VLM (large))
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2021-11-22
63.2
best: 68 (VAST)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2021-11-22
85.7
best: 92.8 (BEiT-3)
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432

Robots4 results

Activity RecognitiononKinetics-600
Top-1 Accuracy· 2021-11-22
87.8
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Activity RecognitiononKinetics-600
Top-5 Accuracy· 2021-11-22
97.8
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Activity RecognitiononKinetics-400
Top-1 Accuracy· 2021-11-22
86.5
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Activity RecognitiononKinetics-400
Top-5 Accuracy· 2021-11-22
97.3
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432

Time Series4 results

Action RecognitiononKinetics-600
Top-1 Accuracy· 2021-11-22
87.8
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Action RecognitiononKinetics-600
Top-5 Accuracy· 2021-11-22
97.8
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Action RecognitiononKinetics-400
Top-1 Accuracy· 2021-11-22
86.5
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432
Action RecognitiononKinetics-400
Top-5 Accuracy· 2021-11-22
97.3
SOTA
Florence: A New Foundation Model for Computer Vision arXiv:2111.11432