This repository collects the latest research progress of Contrastive Learning (CL) and Data Augmentation (DA) in Recommender Systems. Comments and contributions are welcome.
CF = Collaborative Filtering, SSL = Self-Supervised Learning
- Survey/Tutorial Total Papers: 3
- Only Data Augmentation Total Papers: 16
- Graph Models with CL Total Papers: 32
- Sequential Models with CL Total Papers: 22
- Other Tasks with CL Total Papers: 26
-
Self-Supervised Learning for Recommender Systems A Survey (Survey)
-
Self-Supervised Learning in Recommendation: Fundamentals and Advances (Tutorial)
WWW 2022, [Web]
-
Tutorial: Self-Supervised Learning for Recommendation: Foundations, Methods and Prospects (Tutorial)
DASFAA 2023, [Web]
-
Enhancing Collaborative Filtering with Generative Augmentation (CF + GAN + DA)
KDD 2019, [PDF]
-
Future Data Helps Training Modeling Future Contexts for Session-based Recommendation (Session + DA)
-
Augmenting Sequential Recommendation with Pseudo-Prior Items via Reversely Pre-training Transformer (Sequential + DA)
-
Self-Knowledge Distillation with Bidirectional Chronological Augmentation of Transformer for Sequential Recommendation (Sequential + DA)
-
Counterfactual Data-Augmented Sequential Recommendation (Sequential + Counterfactual + DA)
SIGIR 2021, [PDF]
-
CauseRec: Counterfactual User Sequence Synthesis for Sequential Recommendation (Sequential + Counterfactual + DA)
SIGIR 2021, [PDF]
-
Explicit Counterfactual Data Augmentation for Recommendation (Sequential + Counterfactual + DA)
WSDM 2023
-
Effective and Efficient Training for Sequential Recommendation using Recency Sampling (Sequential + DA)
RecSys 2022, [PDF]
-
Data Augmentation Strategies for Improving Sequential Recommender Systems (Sequential + DA)
-
Learning to Augment for Casual User Recommendation (Sequential + DA)
WWW 2022, [PDF]
-
Recency Dropout for Recurrent Recommender Systems (RNN + DA)
arXiv 2022, [PDF]
-
Improved Recurrent Neural Networks for Session-based Recommendations (RNN + DA)
DLRS 2016, [PDF]
-
Bootstrapping User and Item Representations for One-Class Collaborative Filtering (CF + Graph + DA)
-
MixGCF: An Improved Training Method for Graph Neural Network-based Recommender Systems (Graph + DA)
-
Improving Recommendation Fairness via Data Augmentation (Fairness + DA)
-
Fairly Adaptive Negative Sampling for Recommendations (Fairness + DA)
WWW 2023, [PDF]
-
Self-supervised Graph Learning for Recommendation (Graph + CL + DA)
-
Contrastive Graph Structure Learning via Information Bottleneck for Recommendation (Graph + CL)
-
Are graph augmentations necessary? simple graph contrastive learning for recommendation (Graph + CL)
-
XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation (Graph + CL)
-
Intent-aware Multi-source Contrastive Alignment for Tag-enhanced Recommendation (Graph + CL + DA)
arXiv 2022, [PDF]
-
DisenPOI: Disentangling Sequential and Geographical Influence for Point-of-Interest Recommendation (POI Rec, Graph + CL + DA)
-
An MLP-based Algorithm for Efficient Contrastive Graph Recommendations (Short paper, Graph + CL + DA)
SIGIR 2022, [PDF]
-
A Review-aware Graph Contrastive Learning Framework for Recommendation (Graph + CL + DA)
-
Simple Yet Effective Graph Contrastive Learning for Recommendation (Graph + CL + DA)
-
Contrastive Meta Learning with Behavior Multiplicity for Recommendation (Graph + CL + DA)
-
Disentangled Contrastive Learning for Social Recommendation (Short paper, Graph + CL + DA)
CIKM 2022, [PDF]
-
Improving Knowledge-aware Recommendation with Multi-level Interactive Contrastive Learning (Graph + CL)
-
Multi-level Cross-view Contrastive Learning for Knowledge-aware Recommender System (Graph + CL)
-
Knowledge Graph Contrastive Learning for Recommendation (Graph + DA + CL)
-
Temporal Knowledge Graph Reasoning with Historical Contrastive Learning (Graph + CL)
-
Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation (Graph + SSL)
-
SAIL: Self-Augmented Graph Contrastive Learning (Graph + CL)
AAAI 2022, [PDF]
-
Predictive and Contrastive: Dual-Auxiliary Learning for Recommendation (Graph + CL)
arXiv 2022, [PDF]
-
Socially-Aware Self-Supervised Tri-Training for Recommendation (Graph + CL)
-
Predictive and Contrastive: Dual-Auxiliary Learning for Recommendation (Graph + CL)
arXiv 2022, [PDF]
-
Multi-Behavior Dynamic Contrastive Learning for Recommendation (Graph + CL)
arXiv 2022, [PDF]
-
Self-Augmented Recommendation with Hypergraph Contrastive Collaborative Filtering (Graph + CL)
-
Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning (Graph + CF + CL)
-
Semi-deterministic and Contrastive Variational Graph Autoencoder for Recommendation (Graph + CL)
-
Hypergraph Contrastive Collaborative Filtering (Graph + CF + CL + DA)
-
Graph Structure Aware Contrastive Knowledge Distillation for Incremental Learning in Recommender Systems (Short paper, Graph + CL)
-
Double-Scale Self-Supervised Hypergraph Learning for Group Recommendation (Group Rec, Graph + CL + DA)
-
Self-Supervised Hypergraph Transformer for Recommender Systems (Graph + SSL)
-
Episodes Discovery Recommendation with Multi-Source Augmentations (Graph + DA + CL)
arXiv 2023, [PDF]
-
Poincaré Heterogeneous Graph Neural Networks for Sequential Recommendation (Graph + Sequential + CL)
TOIS 2023, [PDF]
-
Adversarial Learning Data Augmentation for Graph Contrastive Learning in Recommendation (Graph + DA + CL)
DASFAA 2023, [PDF]
-
SimCGNN: Simple Contrastive Graph Neural Network for Session-based Recommendation (Graph + CL)
arXiv 2023, [PDF]
-
Uniform Sequence Better: Time Interval Aware Data Augmentation for Sequential Recommendation (Sequential + CL + DA)
-
Contrastive Learning for Sequential Recommendation (Sequential + CL + DA)
-
Contrastive Self-supervised Sequential Recommendation with Robust Augmentation (Sequential + CL + DA)
-
Learnable Model Augmentation Self-Supervised Learning for Sequential Recommendation (Sequential + CL + DA)
arXiv 2022, [PDF]
-
S3-Rec: Self-Supervised Learning for Sequential Recommendation with Mutual Information Maximization (Sequential + CL + DA)
-
Contrastive Curriculum Learning for Sequential User Behavior Modeling via Data Augmentation (Sequential + CL + DA)
-
Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation (Sequential + CL + DA)
-
Memory Augmented Multi-Instance Contrastive Predictive Coding for Sequential Recommendation (Sequential + CL + DA)
arXiv 2021, [PDF]
-
Contrastive Learning with Bidirectional Transformers for Sequential Recommendation (Sequential + CL + DA)
-
ContrastVAE: Contrastive Variational AutoEncoder for Sequential Recommendation (Sequential + CL + DA)
-
Temporal Contrastive Pre-Training for Sequential Recommendation (Sequential + CL + DA)
-
Multi-level Contrastive Learning Framework for Sequential Recommendation (Graph + Sequential + CL)
CIKM 2022, [PDF]
-
Equivariant Contrastive Learning for Sequential Recommendation (Sequential + DA + CL)
arXiv 2022, [PDF]
-
Explanation Guided Contrastive Learning for Sequential Recommendation (Sequential + DA + CL)
-
Intent Contrastive Learning for Sequential Recommendation (Sequential + DA + CL)
-
Dual Contrastive Network for Sequential Recommendation (Short paper, Sequential + CL)
SIGIR 2022, [PDF]
-
Dual Contrastive Network for Sequential Recommendation with User and Item-Centric Perspectives (Sequential + CL)
arXiv 2022, [PDF]
-
Enhancing Sequential Recommendation with Graph Contrastive Learning (Sequential + Graph + CL + DA)
-
Disentangling Long and Short-Term Interests for Recommendation (Sequential + Graph + CL)
-
Hyperbolic Hypergraphs for Sequential Recommendation (Sequential + Graph + CL + DA)
-
Mutual Wasserstein Discrepancy Minimization for Sequential Recommendation (Sequential + CL)
-
Dual-interest Factorization-heads Attention for Sequential Recommendation (Sequential + CL)
-
CL4CTR: A Contrastive Learning Framework for CTR Prediction (CTR + CL)
-
CCL4Rec: Contrast over Contrastive Learning for Micro-video Recommendation (Micro-video + CL)
arXiv 2022, [PDF]
-
Re4: Learning to Re-contrast, Re-attend, Re-construct for Multi-interest Recommendation (Multi-interest + CL)
-
Interventional Recommendation with Contrastive Counterfactual Learning for Better Understanding User Preferences (Counterfactual + DA + CL)
arXiv 2022, [PDF]
-
Multi-granularity Item-based Contrastive Recommendation (Industry + CL)
arXiv 2022, [PDF]
-
Improving Micro-video Recommendation via Contrastive Multiple Interests (Short paper, Micro-video + CL)
SIGIR 2022, [PDF]
-
Exploiting Negative Preference in Content-based Music Recommendation with Contrastive Learning (Music Rec + CL)
-
Self-supervised Learning for Large-scale Item Recommendations (Industry + CL + DA)
CIKM 2021, [PDF]
-
CrossCBR: Cross-view Contrastive Learning for Bundle Recommendation (Bundle Rec + CL)
-
Contrastive Learning for Cold-start Recommendation (Short paper, Cold Start + CL)
ACM MM (ACM International Conference on Multimedia) 2021, [PDF], [Code]
-
Socially-aware Dual Contrastive Learning for Cold-Start Recommendation (Short paper, Cold Start + CL)
SIGIR 2022, [PDF]
-
Multi-modal Graph Contrastive Learning for Micro-video Recommendation (Short paper, Cold Start + Graph + CL)
SIGIR 2022, [PDF]
-
Self-supervised Learning for Multimedia Recommendation (Multimedia Rec + Graph + DA + CL)
-
SelfCF: A Simple Framework for Self-supervised Collaborative Filtering (CF + Graph + DA + CL)
ACM MM (ACM International Conference on Multimedia) 2021, [PDF], [Code]
-
Trading Hard Negatives and True Negatives:A Debiased Contrastive Collaborative Filtering Approach (CF + CL)
IJCAI 2022, [PDF]
-
The World is Binary: Contrastive Learning for Denoising Next Basket Recommendation (Next Basket + CL)
SIGIR 2021, [PDF]
-
MIC: Model-agnostic Integrated Cross-channel Recommender (Industry + CL + DA)
CIKM 2022, [PDF]
-
A Contrastive Sharing Model for Multi-Task Recommendation (Multi-Task + CL)
WWW 2022, [PDF]
-
C2-CRS: Coarse-to-Fine Contrastive Learning for Conversational Recommender System (Conversational Rec + CL)
-
Contrastive Cross-domain Recommendation in Matching (Cross-domain Rec + DA + CL)
-
Contrastive Cross-domain Recommendation in Matching (Cross-domain Rec + Sequential + CL)
-
Prototypical Contrastive Learning and Adaptive Interest Selection for Candidate Generation in Recommendations (Short Paper, Industry + CL + DA)
-
Spatio-Temporal Contrastive Learning Enhanced GNNs for Session-based Recommendation (GNN + CL)
TOIS 2022, under review, [PDF]
-
Unsupervised Legal Evidence Retrieval via Contrastive Learning with Approximate Aggregated Positive (CL)
AAAI 2023
-
Disentangled Causal Embedding With Contrastive Learning For Recommender System (Causal + CL)
-
Contrastive Collaborative Filtering for Cold-Start Item Recommendation (CF + Cold Start + CL)