Gaussian Mixture Flow Matching with Domain Alignment for Multi-Domain Sequential Recommendation

2025-10-23

AI Summary

To address domain heterogeneity, data imbalance, and challenges in modeling cross-domain transfer and scaling in multi-domain sequential recommendation, this paper proposes GMFlowRec, a generative framework. Methodologically, it (1) employs Gaussian Mixture Flow Matching to model sparse and high-frequency cross-domain behavioral trajectories; (2) introduces a domain-alignment prior mechanism coupled with a dual-masked Transformer to jointly capture domain-invariant user intents and domain-specific preferences; and (3) adopts a unified backbone-based generative paradigm to balance expressive power and deployment efficiency. Extensive experiments on multi-domain datasets from JD.com and Amazon demonstrate that GMFlowRec achieves up to a 44% improvement in NDCG@5 over state-of-the-art methods. It exhibits strong generalization across domains and superior scalability, making it suitable for real-world industrial deployment.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Data Mining & Knowledge Management: Recommender Systems
• Application Domains: Humanities & Computational Social Science

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Users increasingly interact with content across multiple domains, resulting in sequential behaviors marked by frequent and complex transitions. While Cross-Domain Sequential Recommendation (CDSR) models two-domain interactions, Multi-Domain Sequential Recommendation (MDSR) introduces significantly more domain transitions, compounded by challenges such as domain heterogeneity and imbalance. Existing approaches often overlook the intricacies of domain transitions, tend to overfit to dense domains while underfitting sparse ones, and struggle to scale effectively as the number of domains increases. We propose extit{GMFlowRec}, an efficient generative framework for MDSR that models domain-aware transition trajectories via Gaussian Mixture Flow Matching. GMFlowRec integrates: (1) a unified dual-masked Transformer to disentangle domain-invariant and domain-specific intents, (2) a Gaussian Mixture flow field to capture diverse behavioral patterns, and (3) a domain-aligned prior to support frequent and sparse transitions. Extensive experiments on JD and Amazon datasets demonstrate that GMFlowRec achieves state-of-the-art performance with up to 44% improvement in NDCG@5, while maintaining high efficiency via a single unified backbone, making it scalable for real-world multi-domain sequential recommendation.

Problem

• Modeling complex cross-domain transitions in sequential recommendation
• Addressing domain heterogeneity and data imbalance issues
• Scaling recommendation systems for multiple domain interactions

Innovation

• Unified dual-masked Transformer for intent disentanglement
• Gaussian Mixture flow field capturing behavioral patterns
• Domain-aligned prior supporting sparse transitions