PCGNet: Unifying Shared and Specific Information for Fashion Matching Recommendations
Abstract
In fashion domain, recommending complementary clothing items that match selected pieces is a crucial cross-selling technique that improves customer satisfaction. Nevertheless, fashion matching presents significant challenges, as recommendations must not only align with individual user fashion preferences but also ensure compatibility between garments. These challenges are twofold. First, existing models often assume an overly simplified decoupled relationship between product compatibility and personalized user preferences, overlooking the natural complexity between the two. Second, existing data-driven approaches are not optimized for real-world fashion data, which is typically sparse and characterized by noisy interactions. To address these challenges, we propose Personalized Compatibility Graph Network, a multi-objective graph learning framework that organically unifies the modeling of product compatibility and personal preferences. PCGNet uses contrastive mutual information maximization to extract and align shared and view-specific patterns, thereby capturing the complex interplay between compatibility and personal preferences. Moreover, we introduce a correlation-aware neighbor sampling and a learnable global graph augmentation, which enhance the model by incorporating self-supervised signals mined directly from the graph, ensuring more stable and informative representations. Finally, PCGNet generates recommendation scores through the joint optimization of BPR ranking loss and multi-view mutual information losses. Experimental validation on two benchmark datasets demonstrates that PCGNet significantly outperforming state-of-the-art methods across all four evaluation metrics.
Related
- A unifying theory for the evaluation of conditional Akaike information for mixed-effects models
- Unifying Privacy Accounting: Information Equivalence and Information Loss
- The mechanism of fashion consciousness influencing Gen Z’s Willingness to continue using slow fashion apparel in a sustainable context
- Deep Learning-Based Object Detection for Fashion, Uniform, and PPE Items: A Systematic Literature Review
- Functional-Specific Information Borrowing under Limited Overlap
BibTeX
@misc{liao2026pcgnet,
title = {{PCGNet: Unifying Shared and Specific Information for Fashion Matching Recommendations}},
author = {Shuiying Liao and P. Y. Mok},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.13339},
eprint = {2609.13339},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.13339},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - PCGNet: Unifying Shared and Specific Information for Fashion Matching Recommendations
AU - Shuiying Liao
AU - P. Y. Mok
PY - 2026
DA - 2026/09/11
PB - arXiv
DO - 10.48550/arxiv.2609.13339
UR - https://arxiv.org/abs/2609.13339
AB - In fashion domain, recommending complementary clothing items that match selected pieces is a crucial cross-selling technique that improves customer satisfaction. Nevertheless, fashion matching presents significant challenges, as recommendations must not only align with individual user fashion preferences but also ensure compatibility between garments. These challenges are twofold. First, existing models often assume an overly simplified decoupled relationship between product compatibility and personalized user preferences, overlooking the natural complexity between the two. Second, existing data-driven approaches are not optimized for real-world fashion data, which is typically sparse and characterized by noisy interactions. To address these challenges, we propose Personalized Compatibility Graph Network, a multi-objective graph learning framework that organically unifies the modeling of product compatibility and personal preferences. PCGNet uses contrastive mutual information maximization to extract and align shared and view-specific patterns, thereby capturing the complex interplay between compatibility and personal preferences. Moreover, we introduce a correlation-aware neighbor sampling and a learnable global graph augmentation, which enhance the model by incorporating self-supervised signals mined directly from the graph, ensuring more stable and informative representations. Finally, PCGNet generates recommendation scores through the joint optimization of BPR ranking loss and multi-view mutual information losses. Experimental validation on two benchmark datasets demonstrates that PCGNet significantly outperforming state-of-the-art methods across all four evaluation metrics.
ER -