PBiLoss is a model-agnostic regularization loss with PopPos and PopNeg sampling that reduces popularity bias metrics PRU and PRI by up to 10% in GNN recommenders while preserving accuracy on datasets like MovieLens.
Heterophily-aware fair recommendation using graph convolu- tional networks
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.IR 3years
2025 3verdicts
UNVERDICTED 3representative citing papers
GNN recommender uses edge classification and cost-sensitive learning to disentangle popularity bias from quality, reporting ~32% average fairness gains with competitive accuracy.
ALDA4Rec improves sequential recommendation by denoising item-item graphs via community detection and adaptively fusing short-term GCN embeddings with long-term sequence models using GRUs, attention, and MLP weighting.
citing papers explorer
-
PBiLoss: Popularity-Aware Regularization to Improve Fairness in Graph-Based Recommender Systems
PBiLoss is a model-agnostic regularization loss with PopPos and PopNeg sampling that reduces popularity bias metrics PRU and PRI by up to 10% in GNN recommenders while preserving accuracy on datasets like MovieLens.
-
Disentangling Popularity and Quality: An Edge Classification Approach for Fair Recommendation
GNN recommender uses edge classification and cost-sensitive learning to disentangle popularity bias from quality, reporting ~32% average fairness gains with competitive accuracy.
-
Adaptive Long-term Embedding with Denoising and Augmentation for Recommendation
ALDA4Rec improves sequential recommendation by denoising item-item graphs via community detection and adaptively fusing short-term GCN embeddings with long-term sequence models using GRUs, attention, and MLP weighting.