Data to 7 Oct 2026

Prototypical Contrastive Learning of Unsupervised Representations

arXiv 2005.04966 ↗First posted 11 May 2020cs.CV cs.LGcited by 636

Not on the map. Why?

Abstract (arXiv)

This paper presents Prototypical Contrastive Learning (PCL), an unsupervised representation learning method that addresses the fundamental limitations of instance-wise contrastive learning. PCL not only learns low-level features for the task of instance discrimination, but more importantly, it implicitly encodes semantic structures of the data into the learned embedding space. Specifically, we introduce prototypes as latent variables to help find the maximum-likelihood estimation of the network parameters in an Expectation-Maximization framework. We iteratively perform E-step as finding the distribution of prototypes via clustering and M-step as optimizing the network via contrastive learning. We propose ProtoNCE loss, a generalized version of the InfoNCE loss for contrastive learning, which encourages representations to be closer to their assigned prototypes. PCL outperforms state-of-the-art instance-wise contrastive learning methods on multiple benchmarks with substantial improvement in low-resource transfer learning. Code and pretrained models are available at https://github.com/salesforce/PCL.

Key facts. Cited by 636 papers in the map (data through 7 Oct 2026). First posted on arXiv 11 May 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 118; 2022: 165; 2023: 150; 2024: 98; 2025: 68; 2026: 37.

636 papers.

Sort:

By area:

2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,848
2021BEiT: BERT Pre-Training of Image Transformerscited by 2,360
2021Align before Fuse: Vision and Language Representation Learning with Momentum Distillationcited by 1,612
2021VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learningcited by 1,013
2021Understanding Dimensional Collapse in Contrastive Self-supervised Learningcited by 374
2021With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representationscited by 324
2022Context Autoencoder for Self-Supervised Representation Learningcited by 297
2023A Cookbook of Self-Supervised Learningcited by 248
2023A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trendscited by 205
2021Self-supervised Pretraining of Visual Features in the Wildcited by 197
2021Unsupervised Semantic Segmentation by Contrasting Object Mask Proposalscited by 190

38 more

2021BEVT: BERT Pretraining of Video Transformerscited by 169
2023TEST: Text Prototype Aligned Embedding to Activate LLM's Ability for Time Seriescited by 165
2021On Feature Decorrelation in Self-Supervised Learningcited by 160
2022Multi-Granularity Cross-modal Alignment for Generalized Medical Visual Representation Learningcited by 159
2022Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localizationcited by 150
2021Efficient Self-supervised Vision Transformers for Representation Learningcited by 149
2022Balanced Contrastive Learning for Long-Tailed Visual Recognitioncited by 143
2022Rethinking Semantic Segmentation: A Prototype Viewcited by 140
2022Self-supervised Learning in Remote Sensing: A Reviewcited by 137
2021SEED: Self-supervised Distillation For Visual Representationcited by 132
2022How to Exploit Hyperspherical Embeddings for Out-of-Distribution Detection?cited by 128
2021Self-supervised Graph-level Representation Learning with Local and Global Structurecited by 125
2022SimMatch: Semi-supervised Learning with Similarity Matchingcited by 118
2022What to Hide from Your Students: Attention-Guided Masked Image Modelingcited by 117
2022PMR: Prototypical Modal Rebalance for Multimodal Learningcited by 114
2021Contrastive Learning with Stronger Augmentationscited by 100
2021SiT: Self-supervised vIsion Transformercited by 99
2022OmniMAE: Single Model Masked Pretraining on Images and Videoscited by 99
2021Large-scale Unsupervised Semantic Segmentationcited by 98
2023Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospectscited by 97
2023Personalized Federated Learning with Feature Alignment and Classifier Collaborationcited by 97
2022Attracting and Dispersing: A Simple Approach for Source-free Domain Adaptationcited by 93
2021Supporting Clustering with Contrastive Learningcited by 90
2022Intent Contrastive Learning for Sequential Recommendationcited by 90
2023Dynamic Conceptional Contrastive Learning for Generalized Category Discoverycited by 90
2021CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIPcited by 88
2022Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learningcited by 86
2021Divide and Contrast: Self-supervised Learning from Uncurated Datacited by 85
2021Delving Deep into the Generalization of Vision Transformers under Distribution Shiftscited by 85
2021Weakly Supervised Semantic Segmentation by Pixel-to-Prototype Contrastcited by 85
2021Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptationcited by 83
2021ReSSL: Relational Self-Supervised Learning with Weak Augmentationcited by 83
2021Weakly Supervised Contrastive Learningcited by 81
2021Joint Learning of Localized Representations from Medical Images and Reportscited by 81
2021FedProc: Prototypical Contrastive Federated Learning on Non-IID datacited by 80
2022Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervisioncited by 80
2021Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learningcited by 78
2023OmniSeg3D: Omniversal 3D Segmentation via Hierarchical Contrastive Learningcited by 75

50 most cited of 636 All 636 (JSON)