A Self-Supervised Mixture-of-Experts Framework for Multi-behavior Recommendation

2025-08-26

AI Summary

In multi-behavior recommendation, models exhibit significant performance imbalance between visited (e.g., clicked, added-to-cart) and unvisited items, as a single architecture struggles to jointly optimize both. To address this, we propose MEMBER, a Mixture-of-Experts framework that uniquely integrates self-supervised learning with task-aware expert specialization: dedicated expert subnetworks are designed for visited and unvisited items respectively, and contrastive learning-driven self-supervised signals explicitly optimize their distinct representations. Built upon multi-behavior sequence modeling, MEMBER enables synergistic performance gains for both item categories. Extensive experiments demonstrate that MEMBER consistently outperforms state-of-the-art methods across multiple benchmarks, achieving up to a 65.46% improvement in Hit Ratio@20. Crucially, it substantially narrows the performance gap between visited and unvisited items, validating its capability for balanced representation learning and strong generalization.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

In e-commerce, where users face a vast array of possible item choices, recommender systems are vital for helping them discover suitable items they might otherwise overlook. While many recommender systems primarily rely on a user's purchase history, recent multi-behavior recommender systems incorporate various auxiliary user behaviors, such as item clicks and cart additions, to enhance recommendations. Despite their overall performance gains, their effectiveness varies considerably between visited items (i.e., those a user has interacted with through auxiliary behaviors) and unvisited items (i.e., those with which the user has had no such interactions). Specifically, our analysis reveals that (1) existing multi-behavior recommender systems exhibit a significant gap in recommendation quality between the two item types (visited and unvisited items) and (2) achieving strong performance on both types with a single model architecture remains challenging. To tackle these issues, we propose a novel multi-behavior recommender system, MEMBER. It employs a mixture-of-experts framework, with experts designed to recommend the two item types, respectively. Each expert is trained using a self-supervised method specialized for its design goal. In our comprehensive experiments, we show the effectiveness of MEMBER across both item types, achieving up to 65.46% performance gain over the best competitor in terms of Hit Ratio@20.

Problem

• Addresses performance gap between visited and unvisited items
• Improves multi-behavior recommendation with mixture-of-experts
• Enhances recommendation quality using self-supervised training methods

Innovation

• Mixture-of-experts framework for multi-behavior recommendation
• Self-supervised training method for specialized experts
• Dual expert design for visited and unvisited items