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/GAN

GAN

Reported on 57 benchmarks across 11 tasks · 6 papers · 17 SOTA

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

Computer Vision21 results

Person Re-IdentificationonMarket-1501
Rank-1· 2017-01-26
83.97
best: 98 (st-ReID(RE, RK))
SOTA
Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro arXiv:1701.07717
Person Re-IdentificationonMarket-1501
mAP· 2017-01-26
66.07
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
SOTA
Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro arXiv:1701.07717
Image ClassificationonCIFAR-10, 4000 Labels
Percentage error· 2016-06-10
15.59
best: 3.96 (SimMatch)
SOTA
Improved Techniques for Training GANs arXiv:1606.03498
Image ClassificationonSVHN, 1000 labels
Accuracy· 2016-06-10
91.89
best: 97.58 (EnAET)
SOTA
Improved Techniques for Training GANs arXiv:1606.03498
Semi-Supervised Image ClassificationonCIFAR-10, 4000 Labels
Percentage error· 2016-06-10
15.59
best: 3.96 (SimMatch)
SOTA
Improved Techniques for Training GANs arXiv:1606.03498
Semi-Supervised Image ClassificationonSVHN, 1000 labels
Accuracy· 2016-06-10
91.89
best: 97.58 (EnAET)
SOTA
Improved Techniques for Training GANs arXiv:1606.03498
Image ClusteringonCIFAR-10
ARI· uses extra data· 2015-11-19
0.176
best: 0.989 (TURTLE (CLIP + DINOv2))
SOTA
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonCIFAR-10
Accuracy· uses extra data· 2015-11-19
0.315
best: 0.995 (TURTLE (CLIP + DINOv2))
SOTA
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonCIFAR-10
NMI· uses extra data· 2015-11-19
0.265
best: 0.985 (TURTLE (CLIP + DINOv2))
SOTA
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonTiny-ImageNet
Accuracy· 2015-11-19
0.041
best: 0.698 (PRO-DSC)
SOTA
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonTiny-ImageNet
NMI· 2015-11-19
0.135
best: 0.8178 (ITAE)
SOTA
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Person Re-IdentificationonDukeMTMC-reID
Rank-1· 2017-01-26
67.68
best: 95.6 (CTL Model (ResNet50, 256x128))
Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro arXiv:1701.07717
Person Re-IdentificationonDukeMTMC-reID
mAP· 2017-01-26
47.13
best: 97.1 (DenseIL)
Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro arXiv:1701.07717
Image ClusteringonImageNet-10
Accuracy· 2015-11-19
0.346
best: 0.992 (TAC)
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonImageNet-10
NMI· 2015-11-19
0.225
best: 0.985 (TAC)
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonCIFAR-100
Accuracy· uses extra data· 2015-11-19
0.151
best: 0.898 (TURTLE (CLIP + DINOv2))
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonCIFAR-100
NMI· uses extra data· 2015-11-19
0.12
best: 0.915 (TURTLE (CLIP + DINOv2))
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonSTL-10
Accuracy· uses extra data· 2015-11-19
0.298
best: 0.997 (TURTLE (CLIP + DINOv2))
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonSTL-10
NMI· uses extra data· 2015-11-19
0.21
best: 0.993 (TURTLE (CLIP + DINOv2))
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonImagenet-dog-15
Accuracy· 2015-11-19
0.174
best: 0.943 (MAE-CT (best))
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434
Image ClusteringonImagenet-dog-15
NMI· 2015-11-19
0.121
best: 0.904 (MAE-CT (best))
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv:1511.06434

Natural Language Processing21 results

Data-to-Text GenerationonVIST
BLEU-1· 2018-04-24
62.8
best: 69 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Data-to-Text GenerationonVIST
BLEU-2· 2018-04-24
38.8
best: 44 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Data-to-Text GenerationonVIST
BLEU-3· 2018-04-24
23
best: 25.3 (CoVS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Data-to-Text GenerationonVIST
BLEU-4· 2018-04-24
14
best: 16.7 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Data-to-Text GenerationonVIST
CIDEr· 2018-04-24
9
best: 14.1 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Data-to-Text GenerationonVIST
METEOR· 2018-04-24
35
best: 37.8 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Data-to-Text GenerationonVIST
ROUGE-L· 2018-04-24
29.5
best: 33.1 (TAPM)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
BLEU-1· 2018-04-24
62.8
best: 69 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
BLEU-2· 2018-04-24
38.8
best: 44 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
BLEU-3· 2018-04-24
23
best: 25.3 (CoVS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
BLEU-4· 2018-04-24
14
best: 16.7 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
CIDEr· 2018-04-24
9
best: 14.1 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
METEOR· 2018-04-24
35
best: 37.8 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Visual StorytellingonVIST
ROUGE-L· 2018-04-24
29.5
best: 33.1 (TAPM)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
BLEU-1· 2018-04-24
62.8
best: 69 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
BLEU-2· 2018-04-24
38.8
best: 44 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
BLEU-3· 2018-04-24
23
best: 25.3 (CoVS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
BLEU-4· 2018-04-24
14
best: 16.7 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
CIDEr· 2018-04-24
9
best: 14.1 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
METEOR· 2018-04-24
35
best: 37.8 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Story GenerationonVIST
ROUGE-L· 2018-04-24
29.5
best: 33.1 (TAPM)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160

Adversarial7 results

Text GenerationonVIST
BLEU-1· 2018-04-24
62.8
best: 69 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Text GenerationonVIST
BLEU-2· 2018-04-24
38.8
best: 44 (AOG + ARS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Text GenerationonVIST
BLEU-3· 2018-04-24
23
best: 25.3 (CoVS)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Text GenerationonVIST
BLEU-4· 2018-04-24
14
best: 16.7 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Text GenerationonVIST
CIDEr· 2018-04-24
9
best: 14.1 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Text GenerationonVIST
METEOR· 2018-04-24
35
best: 37.8 (HEGR)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160
Text GenerationonVIST
ROUGE-L· 2018-04-24
29.5
best: 33.1 (TAPM)
No Metrics Are Perfect: Adversarial Reward Learning for Visual Storytelling arXiv:1804.09160

Medical5 results

Image GenerationonCLEVR
FID-5k-training-steps· 2021-03-01
25.0244
best: 0.89 (Projected GAN)
SOTA
Generative Adversarial Transformers arXiv:2103.01209
Image GenerationonFFHQ
FID-10k-training-steps· 2021-03-01
13.1844
best: 10.8309 (StyleGAN2)
SOTA
Generative Adversarial Transformers arXiv:2103.01209
Image GenerationonCityscapes
FID-10k-training-steps· 2021-03-01
11.5652
best: 3.41 (Projected GAN)
SOTA
Generative Adversarial Transformers arXiv:2103.01209
Image GenerationonLSUN Bedroom 256 x 256
FID-10k-training-steps· 2021-03-01
12.1567
best: 1.52 (Projected GAN)
SOTA
Generative Adversarial Transformers arXiv:2103.01209
Synthetic Data GenerationonUCI Epileptic Seizure Recognition
AUROC· uses extra data
0.87
best: 0.92 (corGAN)

Knowledge Base3 results

Text SummarizationonCNN / Daily Mail (Anonymized)
ROUGE-1· 2017-11-26
39.92
best: 42.3 (HSSAS)
SOTA
Generative Adversarial Network for Abstractive Text Summarization arXiv:1711.09357
Text SummarizationonCNN / Daily Mail (Anonymized)
ROUGE-2· 2017-11-26
17.65
best: 18.87 (RNES w/o coherence)
SOTA
Generative Adversarial Network for Abstractive Text Summarization arXiv:1711.09357
Text SummarizationonCNN / Daily Mail (Anonymized)
ROUGE-L· 2017-11-26
36.71
best: 37.75 (RNES w/o coherence)
Generative Adversarial Network for Abstractive Text Summarization arXiv:1711.09357