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/TCL

TCL

Reported on 66 benchmarks across 10 tasks · 4 papers · 24 SOTA

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

Computer Vision28 results

Unsupervised Semantic SegmentationonCOCO-Stuff-171
mIoU· 2022-12-01
22.4
best: 34 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Unsupervised Semantic SegmentationonCOCO-Object
mIoU· 2022-12-01
31.6
best: 49.4 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Unsupervised Semantic SegmentationonADE20K
Mean IoU (val)· 2022-12-01
17.1
best: 30.7 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Unsupervised Semantic SegmentationonPASCAL Context-59
mIoU· 2022-12-01
33.9
best: 50.8 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Unsupervised Semantic SegmentationonPascalVOC-20
mIoU· 2022-12-01
83.2
best: 91.8 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Unsupervised Semantic SegmentationonPASCAL VOC
mIoU· 2022-12-01
55
best: 76.7 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Image ClusteringonSTL-10
ARI· 2022-10-21
0.757
best: 0.994 (TURTLE (CLIP + DINOv2))
SOTA
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image Classificationonmini WebVision 1.0
ImageNet Top-1 Accuracy· 2023-03-13
75.4
best: 82.56 (LRA-diffusion (CLIP ViT))
Twin Contrastive Learning with Noisy Labels arXiv:2303.06930
Image Classificationonmini WebVision 1.0
ImageNet Top-5 Accuracy· 2023-03-13
92.4
best: 97.24 (RTE (Inception-ResNet-v2))
Twin Contrastive Learning with Noisy Labels arXiv:2303.06930
Image Classificationonmini WebVision 1.0
Top-1 Accuracy· 2023-03-13
79.1
best: 84.16 (LRA-diffusion (CLIP ViT))
Twin Contrastive Learning with Noisy Labels arXiv:2303.06930
Image Classificationonmini WebVision 1.0
Top-5 Accuracy· 2023-03-13
92.3
best: 94.84 (PSSCL (130 epochs))
Twin Contrastive Learning with Noisy Labels arXiv:2303.06930
Unsupervised Semantic SegmentationonCityscapes val
mIoU· 2022-12-01
24
best: 51.1 (CorrCLIP)
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Open Vocabulary Semantic SegmentationonPascalVOC-20
mIoU· 2022-12-01
83.2
best: 97.9 (UMG-CLIP-L/14)
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Open Vocabulary Semantic SegmentationonPASCAL Context-59
mIoU· 2022-12-01
33.9
best: 64.6 (HyperSeg)
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Image ClusteringonImageNet-10
ARI· 2022-10-21
0.837
best: 0.935 (DPAC)
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonImageNet-10
Accuracy· 2022-10-21
0.895
best: 0.992 (TAC)
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonImageNet-10
NMI· 2022-10-21
0.875
best: 0.985 (TAC)
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonCIFAR-10
ARI· 2022-10-21
0.78
best: 0.989 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonCIFAR-10
Accuracy· 2022-10-21
0.887
best: 0.995 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonCIFAR-10
NMI· 2022-10-21
0.819
best: 0.985 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonCIFAR-100
ARI· 2022-10-21
0.357
best: 0.834 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonCIFAR-100
Accuracy· 2022-10-21
0.531
best: 0.898 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonCIFAR-100
NMI· 2022-10-21
0.529
best: 0.915 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonSTL-10
Accuracy· 2022-10-21
0.868
best: 0.997 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonSTL-10
NMI· 2022-10-21
0.799
best: 0.993 (TURTLE (CLIP + DINOv2))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonImagenet-dog-15
ARI· 2022-10-21
0.516
best: 0.879 (MAE-CT (best))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonImagenet-dog-15
Accuracy· 2022-10-21
0.644
best: 0.943 (MAE-CT (best))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Image ClusteringonImagenet-dog-15
NMI· 2022-10-21
0.623
best: 0.904 (MAE-CT (best))
Twin Contrastive Learning for Online Clustering arXiv:2210.11680

Miscellaneous18 results

Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· uses extra data· 2022-12-01
83.2
best: 92.8 (BEiT-3)
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2022-12-01
83.2
best: 92.8 (BEiT-3)
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· uses extra data· 2022-02-21
75.6
best: 84.8 (BEiT-3)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@10· uses extra data· 2022-02-21
96.7
best: 98.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· uses extra data· 2022-02-21
92.8
best: 96.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· uses extra data· 2022-02-21
59
best: 68 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· uses extra data· 2022-02-21
89.9
best: 92.8 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· 2022-02-21
71.4
best: 84.8 (BEiT-3)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@10· 2022-02-21
95.4
best: 98.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· 2022-02-21
90.8
best: 96.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· 2022-02-21
53.5
best: 68 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· 2022-02-21
87.1
best: 92.8 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· 2022-02-21
79
best: 92.8 (BEiT-3)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2022-02-21
75.6
best: 84.8 (BEiT-3)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2022-02-21
96.7
best: 98.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2022-02-21
92.8
best: 96.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2022-02-21
59
best: 68 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2022-02-21
89.9
best: 92.8 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401

Natural Language Processing10 results

Text ClusteringonStackoverflow
Acc· 2022-10-21
88.2
SOTA
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Text ClusteringonBiomedical
Acc· 2022-10-21
49.8
SOTA
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Text ClusteringonBiomedical
NMI· 2022-10-21
42.9
SOTA
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2022-12-01
83.2
best: 92.8 (BEiT-3)
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Text ClusteringonStackoverflow
NMI· 2022-10-21
0.786
best: 77.08 (DSSCC)
Twin Contrastive Learning for Online Clustering arXiv:2210.11680
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2022-02-21
75.6
best: 84.8 (BEiT-3)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2022-02-21
96.7
best: 98.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2022-02-21
92.8
best: 96.5 (X2-VLM (large))
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2022-02-21
59
best: 68 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2022-02-21
89.9
best: 92.8 (VAST)
Vision-Language Pre-Training with Triple Contrastive Learning arXiv:2202.10401

Medical8 results

Semantic SegmentationonCC3M-TagMask
mIoU· 2022-12-01
60.4
best: 65.5 (TTD (TCL))
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonCOCO-Stuff-171
mIoU· 2022-12-01
22.4
best: 34 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonCOCO-Object
mIoU· 2022-12-01
31.6
best: 49.4 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonADE20K
Mean IoU (val)· 2022-12-01
17.1
best: 30.7 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonPASCAL Context-59
mIoU· 2022-12-01
33.9
best: 50.8 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonPascalVOC-20
mIoU· 2022-12-01
83.2
best: 91.8 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonPASCAL VOC
mIoU· 2022-12-01
55
best: 76.7 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
Semantic SegmentationonCityscapes val
mIoU· 2022-12-01
24
best: 90.3 (EfficientPS (Cityscapes-fine))
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785

Audio8 results

10-shot image generationonCC3M-TagMask
mIoU· 2022-12-01
60.4
best: 65.5 (TTD (TCL))
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonCOCO-Stuff-171
mIoU· 2022-12-01
22.4
best: 34 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonCOCO-Object
mIoU· 2022-12-01
31.6
best: 49.4 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonADE20K
Mean IoU (val)· 2022-12-01
17.1
best: 30.7 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonPASCAL Context-59
mIoU· 2022-12-01
33.9
best: 50.8 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonPascalVOC-20
mIoU· 2022-12-01
83.2
best: 91.8 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonPASCAL VOC
mIoU· 2022-12-01
55
best: 76.7 (CorrCLIP)
SOTA
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785
10-shot image generationonCityscapes val
mIoU· 2022-12-01
24
best: 90.3 (EfficientPS (Cityscapes-fine))
Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs arXiv:2212.00785