Learning Binarized Representations with Pseudo-positive Sample Enhancement for Efficient Graph Collaborative Filtering

2025/06/03 by Yankai Chen, Chen, Yankai, Xinni Zhang +6
Computer Science · #Advanced Graph Neural Networks #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Recommender Systems and Techniques

paper · pdf · doi:10.48550/arxiv.2506.02750

openalex publication_date 2025/06/03 · openalex created_date 2025/10/14 · openalex updated_date 2026/07/28

Abstract

Learning vectorized embeddings is fundamental to many recommender systems for user-item matching. To enable efficient online inference, representation binarization, which embeds latent features into compact binary sequences, has recently shown significant promise in optimizing both memory usage and computational overhead. However, existing approaches primarily focus on numerical quantization, neglecting the associated information loss, which often results in noticeable performance degradation. To address these issues, we study the problem of graph representation binarization for efficient collaborative filtering. Our findings indicate that explicitly mitigating information loss at various stages of embedding binarization has a significant positive impact on performance. Building on these insights, we propose an enhanced framework, BiGeaR++, which specifically leverages supervisory signals from pseudo-positive samples, incorporating both real item data and latent embedding samples. Compared to its predecessor BiGeaR, BiGeaR++ introduces a fine-grained inference distillation mechanism and an effective embedding sample synthesis approach. Empirical evaluations across five real-world datasets demonstrate that the new designs in BiGeaR++ work seamlessly well with other modules, delivering substantial improvements of around 1%-10% over BiGeaR and thus achieving state-of-the-art performance compared to the competing methods. Our implementation is available at https://github.com/QueYork/BiGeaR-SS.

Citations

• Semi-supervised Node Importance Estimation with Informative Distribution Modeling for Uncertainty Regularization
• Topology-Aware Conformal Prediction for Stream Networks
• Benchmarking Multimodal Retrieval Augmented Generation with Dynamic VQA Dataset and Self-adaptive Planning Agent
• BANGS: Game-Theoretic Node Selection for Graph Self-Training
• UltraWiki: Ultra-fine-grained Entity Set Expansion with Negative Seed Entities
• HiHPQ: Hierarchical Hyperbolic Product Quantization for Unsupervised Image Retrieval
• LLMRec: Large Language Models with Graph Augmentation for Recommendation
• MESED: A Multi-modal Entity Set Expansion Dataset with Fine-grained Semantic Classes and Hard Negative Entities
• Knowledge Distillation on Graphs: A Survey
• Knowledge-aware Neural Networks with Personalized Feature Referencing for Cold-start Recommendation
• Learning Binarized Graph Representations with Multi-faceted Quantization Reinforcement for Top-K Recommendation
• Deep Hash Distillation for Image Retrieval
• Towards Low-loss 1-bit Quantization of User-item Representations for Top-K Recommendation
• Attentive Knowledge-aware Graph Convolutional Networks with Collaborative Guidance for Personalized Recommendation
• Modeling Scale-free Graphs with Hyperbolic Geometry for Knowledge-aware Recommendation
• Distilling Holistic Knowledge with Graph Neural Networks
• Graph-Free Knowledge Distillation for Graph Neural Networks
• Binary Graph Neural Networks
• Learning to Embed Categorical Features without Embedding Tables for Recommendation
• Understanding Negative Sampling in Graph Representation Learning
• Creating Something from Nothing: Unsupervised Knowledge Distillation for Cross-Modal Hashing
• Distilling Knowledge from Graph Convolutional Networks
• LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
• LightGCN
• Self-training with Noisy Student improves ImageNet classification
• Self-Training With Noisy Student Improves ImageNet Classification
• Forward and Backward Information Retention for Accurate Binary Neural Networks
• Candidate Generation with Binary Codes for Large-Scale Top-N Recommendation
• Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks
• RBCN: Rectified Binary Convolutional Networks for Enhancing the Performance of 1-bit DCNNs
• Quantization Networks
• DeepGCNs: Can GCNs Go as Deep as CNNs?
• Simplifying Graph Convolutional Networks
• Regularized Binary Network Training
• Ranking Distillation: Learning Compact Ranking Models With High Performance for Recommender System
• Representation Learning on Graphs with Jumping Knowledge Networks
• Scalable Methods for 8-bit Training of Neural Networks
• Large scale distributed neural network training through online distillation
• Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning
• Graph Attention Networks
• Inductive Representation Learning on Large Graphs
• Neural Collaborative Filtering
• Understanding Black-box Predictions via Influence Functions
• HashNet: Deep Learning to Hash by Continuation
• Categorical Reparameterization with Gumbel-Softmax
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Semi-Supervised Classification with Graph Convolutional Networks
• A Survey on Learning to Hash
• XNOR-Net: ImageNet Classification Using Binary Convolutional Neural\n Networks
• Modeling User Exposure in Recommendation
• Distilling the Knowledge in a Neural Network
• Adam: A Method for Stochastic Optimization
• Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
• Matrix Factorization Techniques for Recommender Systems
• BPR: Bayesian Personalized Ranking from Implicit Feedback
• A Comprehensive Survey on Graph Neural Networks

Related

• Learning Binarized Graph Representations with Multi-faceted Quantization Reinforcement for Top-K Recommendation
• Binarized Collaborative Filtering with Distilling Graph Convolutional Networks
• TwinCL: A Twin Graph Contrastive Learning Model for Collaborative Filtering
• Graph Attention Collaborative Similarity Embedding for Recommender System
• Vectorized Context-Aware Embeddings for GAT-Based Collaborative Filtering