Compositions of Variant Experts for Integrating Short-Term and Long-Term Preferences

2025-06-29

AI Summary

To address the challenge of jointly modeling short-term dynamic interests and long-term stable preferences in personalized sequential recommendation, this paper proposes CoVE—a collaborative multi-expert neural architecture that dynamically fuses multi-scale user behavioral representations via learnable gating mechanisms. Its core innovations include: (i) specialized expert modules explicitly modeling short-term click sequences and long-term interaction histories, and (ii) a collaborative attention mechanism uncovering their interplay. Extensive experiments on four real-world datasets (Amazon, Yelp, etc.) demonstrate that CoVE significantly outperforms state-of-the-art methods, achieving an average 3.7% improvement in Recall@20. Ablation studies confirm the efficacy of both expert specialization and collaborative modeling. This work establishes a novel, interpretable, and scalable paradigm for multi-timescale user behavior modeling.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Web Mining and Content Analysis: Models for Web evolution
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

In the online digital realm, recommendation systems are ubiquitous and play a crucial role in enhancing user experience. These systems leverage user preferences to provide personalized recommendations, thereby helping users navigate through the paradox of choice. This work focuses on personalized sequential recommendation, where the system considers not only a user's immediate, evolving session context, but also their cumulative historical behavior to provide highly relevant and timely recommendations. Through an empirical study conducted on diverse real-world datasets, we have observed and quantified the existence and impact of both short-term (immediate and transient) and long-term (enduring and stable) preferences on users' historical interactions. Building on these insights, we propose a framework that combines short- and long-term preferences to enhance recommendation performance, namely Compositions of Variant Experts (CoVE). This novel framework dynamically integrates short- and long-term preferences through the use of different specialized recommendation models (i.e., experts). Extensive experiments showcase the effectiveness of the proposed methods and ablation studies further investigate the impact of variant expert types.

Problem

• Integrating short-term and long-term user preferences
• Enhancing personalized sequential recommendation systems
• Dynamically combining variant expert models

Innovation

• Dynamic integration of short and long-term preferences
• Specialized recommendation models as variant experts
• Enhanced performance through empirical dataset analysis