Contrastive Representation Learning: A Framework and Review
Open accessThe DOI matches its Crossref record; 147 of the 162 references it cites were found in a registry or an index, and 15 are not checked yet.
Abstract
Contrastive Learning has recently received interest due to its success in self-supervised representation learning in the computer vision domain. However, the origins of Contrastive Learning date as far back as the 1990s and its development has spanned across many fields and domains including Metric Learning and natural language processing. In this paper, we provide a comprehensive literature review and we propose a general Contrastive Representation Learning framework that simplifies and unifies many different contrastive learning methods. We also provide a taxonomy for each of the components of contrastive learning in order to summarise it and distinguish it from other forms of machine learning. We then discuss the inductive biases which are present in any contrastive learning system and we analyse our framework under different views from various sub-fields of Machine Learning. Examples of how contrastive learning has been applied in computer vision, natural language processing, audio processing, and others, as well as in Reinforcement Learning are also presented. Finally, we discuss the challenges and some of the most promising future research directions ahead.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“According to the suggestions of Le-Khac et al. [29], the main methods can be divided into the following categories: multisensory signals, data augmentation, localglobal consistency, and temporal consistency.”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [29]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Contrastive learning [56-58] is in fact a training procedure used in recent artificial intelligence models as a way to efficiently learn semantic representations.”
From How Adversarial REM Dreams May Facilitate Creativity, and Why We Become Aware of Them (2024) · cites this paper as [56]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“As summarized in [29], commonly used contrastive loss functions generally include scoring functions (cosine similarity), energy-based margin functions (pair loss and triplet loss), probabilistic NCE-based functions, and mutual information based functions.”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [29]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 2×
“In SimCLRv2 [17], empirical experiments show that the output of the second layer of a 3-layer MLP projection head is a better representation for transfer learning than the output of the base ResNet [42] encoder.”
BackgroundMethodVaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 3×
“Recently, the Bidirectional Encoder Representation from Transformer (BERT) [24] model learns bidirectional word representations using the Transformer architecture’s decoder [103] and demonstrated great performance for transfer learning in multiple downstream tasks.”
MethodHochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“Even though the BERT [24] text encoder and the Faster RCNN [87] object detection model are trained separately and not updated by the contrastive loss, the LSTM cells [48] and a linear project head can still map between words and object representations.”
MethodExploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173 · cited 5×
“The discretised speech representation can then be used directly as a representation for other models that expect discrete input such as BERT [24].”
MethodInfluentialFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 5×
“The discretised speech representation can then be used directly as a representation for other models that expect discrete input such as BERT [24].”
MethodInfluentialEfficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568
Research works printed but not linked: 1
- stratos, 2019, Noise contrastive estimation
Cited by
Items 401–500 of 948
Ranking Enhanced Supervised Contrastive Learning for Regression
Zhou, Z., Zhao, Y., Zuo, H., & Chen, W. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-2253-2_2 · Cited by 5
Enhancing Low-Cost Molecular Property Prediction with Contrastive Learning on SMILES Representations
Quiles, M. G., Ribeiro, P. A. L., Pinheiro, G. A., Prati, R. C., & Silva, J. L. F. D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-65329-2_26 · Cited by 6
LoFT: LoRA-Based Efficient and Robust Fine-Tuning Framework for Adversarial Training
Fu, J., Fang, J., Sun, J., Zhuang, S., Geng, L., & Liu, Y. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10651480 · Cited by 11
“As a very important part of SSL, contrastive learning can efficiently extract deeper features of unlabeled data based on the similarities and differences between these data [31].”
BackgroundItTakesTwo: Leveraging Peer Representations for Semi-supervised LiDAR Semantic Segmentation
Liu, Y., Chen, Y., Wang, H., Belagiannis, V., Reid, I. D., & Carneiro, G. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73232-4_5 · Cited by 9
Li, D., Zhang, L., Jie, H. J., Xiong, N. N., Zhang, L., & Wan, J. (2024) · Complex & Intelligent Systems · doi:10.1007/s40747-024-01642-6 · Cited by 5
“In the fields of natural language processing and computer vision,contrastivelearningmethodshaveachievedsomesuc-cess (Khac et al. [18]; Yan et al. [19]; Zhang et al. [21]).”
Real-Time Imputation Model for Missing Sensor Data Based on Alternating Attention Mechanism
Zhang, M., Zhao, R., Wang, C., Jing, L., & Li, D. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3519370 · Cited by 5
“Contrastive learning is an unsupervised learning method aimed at obtaining effective feature representations by learning similarities and differences between samples [38].”
Vural, O., Hamdi, S. M., & Boubrahimi, S. F. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10825566 · Cited by 5
DAISY: Dual-Granularity Contrastive Learning for Disease Similarity Prediction
Gao, Z., Ma, H., Wang, Y., Li, Z., & Chang, L. (2025) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3536100 · Cited by 5
“Contrastive learning has achieved significant success in computer vision [29], [30], [31] and natural language processing tasks [32], [33] by maximizing the consistency between different views.”
Feature distribution statistics as a loss objective for robust white balance correction
Kınlı, F., & Kıraç, F. (2025) · Machine Vision and Applications · doi:10.1007/s00138-025-01680-1 · Cited by 5
Zero-shot and degradation aware learning for polarization super-resolution
Zhu, Z., Li, Y., Zhang, J., Zhao, W. F., & Xiong, J. (2025) · Optics & Laser Technology · doi:10.1016/j.optlastec.2025.113036 · Cited by 5
Abdollahpour, N., Kakaraparthi, A., Valluru, C. S., & Artan, N. S. (2025) · IEEE Access · doi:10.1109/access.2025.3576056 · Cited by 8
“A study [70] applied the Contrastive Representation Learning (CRL) [99] method for self-supervised learning.”
InfluentialContrastive Learning for Generating Optical Coherence Tomography Images of the Retina
Kaplan, S., & Lensu, L. T. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16980-9_11 · Cited by 7
Iterative Contrastive Learning for Single Image Raindrop Removal
Yang, X., Qian, P., Li, W., Zhao, S., Chen, C., Li, X., & Zeng, Z. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9897979 · Cited by 5
“Interested readers can refer to [14] for a comprehensive literature review.”
BackgroundSrPPG: Semi-Supervised Adversarial Learning for Remote Photoplethysmography with Noisy Data
Hasan, Z., Faridee, A. Z. M., Ahmed, M., Ayyanar, S., & Roy, N. (2023) · IEEE International Conference on Smart Computing · doi:10.1109/smartcomp58114.2023.00021 · Cited by 5
“We take motivation from the contrastive supervised sampling method [32] and utilize the noisy concurrent yi information as a pivot for sampling video snippets to reduce false sampling.”
MethodRevisiting Low-Resolution Images Retrieval with Attention Mechanism and Contrastive Learning
Vu, D. T., Yu, G.-H., & Kim, J. Y. (2021) · Applied Sciences · doi:10.3390/app11156783 · Cited by 6
Supervised Contrastive Learning for Game-Play Frustration Detection from Speech
Song, M., Parada‐Cabaleiro, E., Liu, S., Milling, M., Baird, A., Yang, Z., & Schuller, B. W. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-78092-0_43 · Cited by 5
“The performance of a deep learning system is directly influenced by the choice and quality of the data representation [24].”
InfluentialComparing Reconstruction- and Contrastive-based Models for Visual Task Planning
Chamzas, C., Lippi, M., Welle, M. C., Varava, A., Kavraki, L. E., & Kragić, D. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981533 · Cited by 6
“Inspired by the revival of contrastive methods in computer vision [14], some recent works rely on contrastive losses to learn efficient state representations.”
MethodMun, S. H., Han, M. H., Lee, D., Kim, J., & Kim, N. S. (2021) · IEEE Access · doi:10.1109/access.2021.3137190 · Cited by 5
“Although the aforementioned methods have shown successful results, the contrastive learning techniques generally require careful handling of the negative pairs, e.g., large batch sizes [13], additional memory banks [14], [22], etc. [10], [11], [15], [23], [24].”
BackgroundMulti-Level Contrastive Learning for Cross-Lingual Alignment
Chen, B., Guo, W., Gu, B., Liu, Q., & Wang, Y. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747720 · Cited by 7
SC-FGCL: Self-Adaptive Cluster-Based Federal Graph Contrastive Learning
Wang, T., Zheng, X., Gao, L., Wan, T., & Tian, L. (2023) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2023.3235593 · Cited by 6
“In recent years, due to the outstanding performance of graph contrastive learning (GCL) in node classification, node clustering and graph classification, it has attracted widespread attention in tasks such as protein structure prediction [1] and computer vision [2].”
BackgroundUnifying the Electron Microscopy Multiverse through a Large-scale Foundation ModelPreprint
He, L., Shi, R., Wang, W., Fang, G., Cai, Y., & Ма, Л. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.13.648639 · Cited by 7
“Image representation learning, often realized through large-scale self-supervised pretraining, aims to extract meaningful and structured information to support multiple downstream analyses 42 .”
Open-Vocabulary Mobile Manipulation Based on Double Relaxed Contrastive Learning With Dense Labeling
Yashima, D., Korekata, R., & Sugiura, K. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3522841 · Cited by 6
“There have been numerous studies in the field of contrastive learning and self-supervised representation learning [14].”
Feng, X., Li, C., & Zhang, S. (2025) · Journal of Consumer Research · doi:10.1093/jcr/ucaf021 · Cited by 6
Zhang, M., Wu, J., Wang, Y., Cao, Y. L., Liu, J., Wang, Q., Song, X., Zhao, J., & Wang, Y. (2025) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2025.145341 · Cited by 4
Reweighting balanced representation learning for long tailed image recognition in multiple domains
Fu, P., Ruhaiyem, N. I. R., & Wang, J. (2025) · Scientific Reports · doi:10.1038/s41598-025-03459-w · Cited by 4
Unifying Graph Contrastive Learning via Graph Message Augmentation
Zhang, Z., Jiang, B., Tang, J., & Luo, B. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3586651 · Cited by 6
“To address this challenge, as a kind of self-supervised learning, contrastive learning [1], [2] has emerged as an important learning method to learn rich graphrepresentationfromunlabeleddata.”
Power System Inertia Estimation Using Contrastive Learning and Ridge Regression
Popovski, P., Chaushevski, A., Acevski, N., & Veljanovski, G. (2025) · International Scientific Conference on Information, Communication and Energy Systems and Technologies (ICEST) · doi:10.1109/icest66328.2025.11098217 · Cited by 4
“Most often its uses include natural language processing, computer vision, audio processing and others [22], [23].”
Kang, G., Ren, H., Liu, J., Cao, B., Xu, Y., & Li, P. (2025) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2025.3602629 · Cited by 4
“This method facilitates the generation of distinct and informative feature representations in an unsupervised fashion, thereby augmenting supervised tasks with unlabeled data [32].”
Zhao, Z., Jiao, Y., Xu, Y., Zhang, X., Zhao, R., & Chen, Z. (2025) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2025.111998 · Cited by 4
Wang, L., Li, Y., & Deng, S. (2025) · Engineering · doi:10.1016/j.eng.2025.08.020 · Cited by 4
Velandia, C., Vidal, Y., & Pozo, F. (2026) · Structural Control and Health Monitoring · doi:10.1155/stc/4023580 · Cited by 4
A Class-Aware Supervised Contrastive Quadratic Neural Network for Imbalanced Bearing Fault Diagnosis
Yu, W.-E., Zhang, S., Sun, J., Li, C., Liao, J.-X., & Zhang, X. (2026) · IEEE Transactions on Reliability · doi:10.1109/tr.2026.3668191 · Cited by 5
“Contrastive learning is a self-supervised representation learning paradigm to mine the latent data representation by exacer-bating the difference among samples of dissimilar labels [28], [39].”
Soft Contrastive Cross-Modal Retrieval
Song, J., Hu, Y., Zhu, L., Zhang, C., Zhang, J., & Zhang, S. (2024) · Applied Sciences · doi:10.3390/app14051944 · Cited by 5
Li, C., Nie, L., Sun, Z., Ding, X., Luo, Q., & Shen, C. (2024) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2024.102394 · Cited by 4
Exploring galaxy properties of eCALIFA with contrastive learning
Martínez-Solaeche, G., García-Benito, R., Delgado, R. M. G., Díaz-García, L. A., Sánchez, S. F., Conrado, A., & Rodríguez-Martín, J. E. (2024) · Astronomy and Astrophysics · doi:10.1051/0004-6361/202450074 · Cited by 6
Integrated self-supervised label propagation for label imbalanced sets
Ge, Z., Yang, Y., & Du, Z. (2024) · Applied Intelligence · doi:10.1007/s10489-024-05591-y · Cited by 4
Feature Augmentation for Self-supervised Contrastive Learning: A Closer Look
Zhang, Y. A., Zhu, R., Zhang, S., Zhou, X. Y., Chen, S., & Chen, X. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10651086 · Cited by 6
“Data augmentation (DA) [23] is the most indispensable part of self-supervised contrastive learning as mentioned in [24].”
BackgroundTaxes are All You Need: Integration Of Taxonomical Hierarchy Relationships Into the Contrastive Loss
Kokilepersaud, K., Yarici, Y., Prabhushankar, M., & AlRegib, G. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10648243 · Cited by 4
Li, H., Qiao, S., & Sun, Y. (2024) · Physics of Fluids · doi:10.1063/5.0237910 · Cited by 4
White, D., Jahangir, M., Mishra, A. K., Baker, C., & Antoniou, M. (2024) · IEEE Transactions on Radar Systems · doi:10.1109/trs.2024.3518842 · Cited by 4
Behzadidoost, R., Izadkhah, H., & Mahan, F. (2024) · Information Sciences · doi:10.1016/j.ins.2024.121756 · Cited by 4
González-Ramírez, A., Atzberger, C., Torres-Román, D., & López, J. (2025) · Remote Sensing · doi:10.3390/rs17030378 · Cited by 4
He, Q., Wang, Y., Du, N., Wang, S., & Song, J. H. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.107635 · Cited by 5
Automated Parts Segmentation of Sperm via a Contrastive Learning-Based Part Matching Network
Chen, W., Song, H., Shan, G., Dai, C., Liu, H., Jiang, A., Sun, C., Ru, C., Librach, C. L., Zhang, Z., et al. (2025) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2025.3567946 · Cited by 4
“Therefore, contrastive learning does not require specific class label for each instance, but only information about which instances are similar (positive pairs) and which are dissimilar (negative pairs) [42].”
Xiao, Z., Luo, C., Hu, J., Sa, G., & Wang, Y. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107859 · Cited by 4
Contrastive learning for unsupervised medical image clustering and reconstruction
Ferrante, M., Boccato, T., Duggento, A., Spasov, S. E., & Toschi, N. (2023) · Proceedings of the Northern Lights Deep Learning Workshop · doi:10.7557/18.6819 · Cited by 5
Cut-in maneuver detection with self-supervised contrastive video representation learning
Nalçakan, Y., & Baştanlar, Y. (2023) · Signal Image and Video Processing · doi:10.1007/s11760-023-02512-3 · Cited by 4
“Self-supervised contrastive learning has recently gained popularity due to its achievements in computer vision [15,16].”
BackgroundPositive and Negative Set Designs in Contrastive Feature Learning for Temporal Action Segmentation
Chen, Y.-C., & Chu, W.-T. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3417392 · Cited by 4
“Additional literature surveys on self-supervised or contrastive learning can be found in [41] and [42].”
Multi-Modal Data-Based Semi-Supervised Learning for Vehicle Positioning
Huan, O., Yang, Y., Luo, T., & Chen, M. (2024) · IEEE Transactions on Communications · doi:10.1109/tcomm.2024.3459848 · Cited by 4
“…as follows: 1) Encoder: The encoder network with parameter ξ E is used to transform each H t,m or H j into a low-dimensional feature vector [33], [34]. f f Specifically, we respectively use z t,m ∈ R 1 × N or z j ∈ R 1 × N to represent the feature vector of each H t,m or H j with N being the…”
MethodSemi-supervised Human Activity Recognition with individual difference alignment
Yang, Z., Li, T., Xu, Z., Huang, Z., Cao, Y., Li, K., & Ma, J. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.126976 · Cited by 5
Suhas, M., Abisset‐Chavanne, E., & Rey, P.-A. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.127513 · Cited by 6
GOTPR: General Outdoor Text-Based Place Recognition Using Scene Graph Retrieval With OpenStreetMap
Jung, D., Kim, K., & Kim, S.-W. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3568306 · Cited by 10
“They embed both the text and pre-segmented point cloud maps using contrastive learning [14], and perform place recognition by finding the most similar embedding pair.”
Peng, T., Feng, C., Danier, D., Zhang, F., Vallade, B. Q. A., Mackin, A., & Bull, D. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-91838-4_7 · Cited by 4
“…quality of enhanced video content is a challenging task and the corresponding labels are difficult to obtain, inspired by contrastive learning [24,31], we employ a projector network (with two layers of multilayer perceptron), g ( · ) , to transform the original task into two classification…”
MethodMartinović, I., Mao, S., Dousty, M., Li, W., Djukanović, M., Colak, E., & Sejdić, E. (2025) · FACETS · doi:10.1139/facets-2024-0229 · Cited by 4
LoopSR: Looping Sim-and-Real for Lifelong Policy Adaptation of Legged Robots
Wu, P., Xie, W., Cao, J., Lai, H., & Zhang, W. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246873 · Cited by 6
Tight PAC-Bayesian Risk Certificates for Contrastive Learning
van Elst, A., & Ghoshdastidar, D. (2025) · SIAM Journal on Mathematics of Data Science · doi:10.1137/24m1715283 · Cited by 7
“Contrastive representation learning is a popular form of self-supervised learning where one aims to extract meaningful features from unlabeled data by distinguishing between similar samples, obtained via augmentations, and dissimilar samples [54, 22, 28].”
Zhang, S., Zou, B., Xu, B., Su, J., & Hu, H. (2021) · IEEE International Conference on Progress in Informatics and Computing · doi:10.1109/pic53636.2021.9687080 · Cited by 5
Encouraging Validatable Features in Machine Learning-Based Highly Automated Driving Functions
Candido, O. D., Köller, M., & Utschick, W. (2022) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2022.3171215 · Cited by 4
“The feature embeddings could additionally be investigated using self-supervised ML methods [54], [55].”
MethodYu-bo, J., Yang, S., Zhou, K., Lu, J., Wang, R. K., Rocliffe, H. R., Pellicoro, A., Cash, J. L., Li, C., & Huang, Z. (2022) · Journal of Biomedical Optics · doi:10.1117/1.jbo.27.8.085002 · Cited by 6
“However, the quality of learned representation depends on the effort of developing effective data augmentation by contrasting latent representation of different augmentations.(35) The most common research is only focused on comparing representation learned by self-supervised methods on pretrained models from ImageNet,…”
BackgroundAkamatsu, Y., Onishi, Y., Imaoka, H., Kameyama, J., & Tsurushima, H. (2022) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3227517 · Cited by 5
“self-supervised or supervised representation learning that differentiates between similar example pairs and dissimilar example pairs [21], [22], [23], [24].”
BackgroundXie, S., Cheng, W., Nie, Z., & Chen, X. (2022) · International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC) · doi:10.1109/sdpc55702.2022.9915840 · Cited by 4
A systematic review for using deep learning in bone scan classification
Kao, Y., Huang, C.-P., Tsai, W.-W., & Yang, J. C. (2023) · Clinical and Translational Imaging · doi:10.1007/s40336-023-00539-7 · Cited by 4
“They also used contrastive representation learning (CRL) [41], a technique that finds implicit similarities between instances without additional labeling, to strengthen acquisitions of high-level representations.”
MethodWho Would be Interested in Services? An Entity Graph Learning System for User Targeting
Yang, D., Hu, B., Yang, X., Shen, Y., Zhang, Z., Gu, J., & Zhang, G. (2023) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde55515.2023.00249 · Cited by 4
“Contrasive learning task Inspired by contrastive learning [13], [14], [36], which works by pulling positive samples closer and pushing negative samples further, we enhance the representation of entities with the auxiliary contrastive supervision.”
BackgroundFederated Representation Learning Through Clustering
Miao, R., & Koyuncu, E. (2023) · Machine learning for signal processing ... · doi:10.1109/mlsp55844.2023.10285898 · Cited by 4
IfCMD: A Novel Method for Radar Target Detection under Complex Clutter Backgrounds
Zhang, C., Xu, Y., Chen, W., Chen, B., Gao, C., & Liu, H. (2024) · Remote Sensing · doi:10.3390/rs16122199 · Cited by 6
Maghouli, M. M., Fereydooni, M., Abdoos, M., & Vahidi-Asl, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.09975 · Cited by 4
“There are several applications for contrastive learning in different domains some of which discussed in [10].”
BackgroundSupSiam: Non-contrastive Auxiliary Loss for Learning from Molecular ConformersPreprint
Maser, M., Park, J. W., Lin, J. Y.-Y., Lee, J. H., Frey, N. C., & Watkins, A. M. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2302.07754 · Cited by 4
“The rationale for this objective states that similar — but different — 2D graphs should map to similar latent vectors (Le-Khac et al., 2020).”
BackgroundUnsupervised contrastive peak caller for ATAC-seq
Vu, H. T. H., Zhang, Y., Tuteja, G., & Dorman, K. S. (2023) · Genome Research · doi:10.1101/gr.277677.123 · Cited by 8
Enhancing Challenging Target Screening via Multimodal Protein-Ligand Contrastive LearningPreprint
Wang, Z., Wang, Z., Yang, M., Pang, L., Nie, F., Liu, S., Gao, Z., Zhao, G., Ji, X., Huang, D., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.08.22.609123 · Cited by 8
“Contrastive learning, recognized as a self-supervised learning paradigm, has achieved notable success in the domains of computer vision and natural language pro-66 cessing [20, 21].”
BackgroundLearning Actionable World Models for Industrial Process ControlPreprint
Yan, P., Abdulkadir, A., Schatte, G. A., Aguzzi, G., Gha, J., Pascher, N., Rosenthal, M., Gao, Y., Grewe, B. F., & Stadelmann, T. (2025) · arXiv (Cornell University) · doi:10.21256/zhaw-33031 · Cited by 6
How Adversarial REM Dreams May Facilitate Creativity, and Why We Become Aware of Them
Deperrois, N., Petrovici, M., Jordan, J., Huber, L., & Senn, W. (2024) · Clinical and Translational Neuroscience · doi:10.3390/ctn8020021 · Cited by 3
Scene Retrieval in Traffic Videos with Contrastive Multimodal Learning
Sadiq, T., & Omlin, C. W. P. (2023) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai59109.2023.00153 · Cited by 4
“It uses contrastive losses to make representations of text and image inputs similar when they correspond to semantically related pairs and dissimilar otherwise [26].”
BackgroundAsymmetric slack contrastive learning for full use of feature information in image translation
Zhang, Y., Li, M., Gou, Y., & He, Y. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112136 · Cited by 3
A Flat-Span Contrastive Learning Method for Nested Named Entity Recognition
Liu, Y., Zhang, K., Tong, R., Cai, C., Chen, D., & Wu, X. (2024) · International Conference on Asian Language Processing · doi:10.1109/ialp63756.2024.10661137 · Cited by 3
“Candidate span classification based on contrastive learning .”
BackgroundA review of lightweight convolutional neural networks for ultrasound signal classification
Zhang, B., Li, Z., Hao, Y., Wang, L., Li, X., & Yao, Y. (2025) · Frontiers in Physiology · doi:10.3389/fphys.2025.1536542 · Cited by 5
Romanov, A., Kurtukova, A., Fedotova, A., & Shelupanov, A. A. (2025) · IEEE Access · doi:10.1109/access.2025.3584616 · Cited by 3
“This model is SimNN [37], [38].”
Enhancing multi-view deep image clustering via contrastive learning for global and local consistency
Shi, F., & Lu, H. (2025) · Pattern Analysis and Applications · doi:10.1007/s10044-025-01538-2 · Cited by 3
“This method [12 ] has been demonstrated to enhance the efficacy of multi-view clustering and representation learning.”
Predicting Depression Treatment Outcome Using Daily Step Count Sensory Data
Hossain, M. Z., Sahoo, S., Shende, C., Patel, P., Morillo, R., Pan, Z., Wang, X., Bi, J., Kamath, J., Russell, A. C., et al. (2025) · ACM Transactions on Computing for Healthcare · doi:10.1145/3777480 · Cited by 4
Huang, X., Zhang, Z., Yang, Z., Ren, C., Yau, D. K. Y., & Deng, R. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3636270 · Cited by 3
“Contrastive learning learns the deep representation of the data characteristics by minimizing the distance between similar pairs and maximizing the distance between dissimilar pairs [15].”
INTENT: Trajectory Prediction Framework With Intention-Guided Contrastive Clustering
Tang, Y., & Ma, W. (2026) · IEEE Open Journal of Intelligent Transportation Systems · doi:10.1109/ojits.2026.3654451 · Cited by 5
“The intention of a road agent can be defined as its subjective tendency to take a specific action or to reach a specific location, which is often seen as the internal reason for taking the corresponding action [33].”
Graph contrastive learning view construction methods in recommender systems: a survey
Yi, Z., Wang, H., Chen, F., Xu, Z., & Yang, J. (2026) · Frontiers of Computer Science · doi:10.1007/s11704-025-50044-5 · Cited by 5
Foundation models in computational pathology: methods, applications and clinical implications
Sali, R., Aschenaki, Y., Leveillee, R. J., Baba, F., Tessema, L., Dixon, C. M., Bachhamba, K., & Zhang, D. Y. (2026) · BMJ Oncology · doi:10.1136/bmjonc-2026-001102 · Cited by 8
Few-label learning for fault diagnosis based on contrastive representations
Peng, D., Yang, Z., Huang, Y., Long, J., Li, C., & Zhang, R. (2022) · International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD) · doi:10.1109/icsmd57530.2022.10058299 · Cited by 3
“This contrastive encoding network is based on a siamese network with contrastive learning loss [6] and consists of an Encoder and a predictor.”
BackgroundSpatiotemporal self-supervised pre-training on satellite imagery improves food insecurity prediction
Cartuyvels, R., Fierens, T., Coppieters, E., Moens, M., & Sileo, D. (2023) · Environmental Data Science · doi:10.1017/eds.2023.42 · Cited by 3
Text Attribute Control via Closed-Loop Disentanglement
Sha, L., & Lukasiewicz, T. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00640 · Cited by 6
Contrastive Consistency and Attentive Complementarity for Deep Multi-View Subspace Clustering
Wang, J., Wu, B., & Zhang, H. (2024) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2023.046011 · Cited by 3
Hierarchical Query Classification in E-commerce Search
He, B., Nag, S., Cui, L., Wang, S., Li, Z., Goutam, R., Li, Z., & Zhang, H. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589335.3648332 · Cited by 3
“The effectiveness of this approach has been showcased in numerous applications, such as image, speech recognition, and natural language processing [21].”
BackgroundContrastive Disentangled Representation Learning for Debiasing Recommendation with Uniform Data
Yang, X., Liu, Z., Lu, X., Yuan, Y., Lu, S., & Gao, Y. (2024) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3627673.3679889 · Cited by 5
“Since uniform data is valuable and small amount, this paper exploit contrastive learning [4, 6] to expand different sample spaces during the training process.”
Contrastive learning for body gesture detection during Adapted Physical Activity
Rocha, J. M., Pennino, F., Dubois, C., Monacelli, É., & Gabbrielli, M. (2024) · IEEE RO-MAN · doi:10.1109/ro-man60168.2024.10731183 · Cited by 3
“This paper investigates contrastive learning [20], an advancement that allows gesture recognition models to learn effective representations from unlabeled data.”
BackgroundRelieving popularity bias in recommender systems via user group-level augmentation
He, M., Zhang, Z., Zhang, H., & Liu, C. (2024) · Applied Soft Computing · doi:10.1016/j.asoc.2024.112610 · Cited by 3
Huang, C., Cai, J., Luo, S., Wang, S., Yang, G., Lei, H., & Zhou, L. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109880 · Cited by 3
A contrast enhanced representation normalization approach to knowledge distillation
Bao, Z., Zhu, D., Du, L., & Li, Y. (2025) · Scientific Reports · doi:10.1038/s41598-025-97699-5 · Cited by 3
Li, C., Wang, K., Gan, H., Zhou, R., Cheng, X., & Yang, Z. (2025) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2025.102590 · Cited by 4
EEG data alignment across devices using a neural network
Olsen, P., Damgaard, A. T., Banluesombatkul, N., & Mikkelsen, K. B. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11254069 · Cited by 3
“This setup shares similarities with simple implementations of contrastive learning [14], as it involves comparing pairs of inputs to learn a meaningful distinction.”
Noise-Robust Representation for Fault Identification with Limited Data via Data Augmentation
Taghiyarrenani, Z., & Berenji, A. (2022) · PHM Society European Conference · doi:10.36001/phme.2022.v7i1.3334 · Cited by 3
Hard Negative Sample Mining for Contrastive Representation in Reinforcement Learning
Chen, Q., Liang, D., & Liu, Y. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-05936-0_22 · Cited by 3
A Note on Representational Understanding
Jakovác, A., & Telcs, A. (2022) · Entropy · doi:10.3390/e24091313 · Cited by 3
“The authors in [1] argue, following the strategy of representation learning [2,3], that understanding a topic is equivalent to finding the right representation of the data.”
BackgroundInfluentialFundamental Concepts and Algorithms
Simeone, O. (2022) · Cambridge University Press eBooks · doi:10.1017/9781009072205.005 · Cited by 3
Learning Generalized Hybrid Proximity Representation for Image Recognition
Li, Z., & Ralescu, A. (2022) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai56018.2022.00138 · Cited by 3
“In supervised metric learning, the model learns to pull together the samples from the same classes and push away the samples from different classes [9].”
BackgroundWang, J., Chen, Z., Zhang, Y., He, D., & Lin, F. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126299 · Cited by 3
Metadata
- Authors
- Le-Khac, P. H., Healy, G., & Smeaton, A. F.
- Affiliations
- Dublin City University (IE)
- Venue
- IEEE Access
- Published
- 2020
- Volume
- 8
- Pages
- 193907-193934
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W3096655658
- Cited by
- 862
- References
- 144
- Citations by year
- 2026: 111 · 2025: 197 · 2024: 212 · 2023: 157 · 2022: 113 · 2021: 71 · 2020: 1
- FWCI
- 33.61
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- arXiv (Cornell University) (acceptedVersion)
- Dublin City University Open Access Institutional Repository (Dublin City University) (submittedVersion, cc-by)
- Arrow@dit (Dublin Institute of Technology) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University) (cc-by)
- Funders
- Science Foundation Ireland through the Science Foundation Ireland (SFI) Centre for Research Training in Machine Learning (18/CRT/6183)
- Insight Centre for Data Analytics (SFI/12/RC/2289_P2)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
- Keywords
- contrastive learning, representation learning, self-supervised learning, metric learning, inductive bias, computer vision, natural language processing, audio signal processing, reinforcement learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.