Multi-Interest Recommendation: A Survey

2025-06-18

AI Summary

Traditional recommender systems struggle to model user preference diversity and item attribute uncertainty. To address this challenge, this paper presents a systematic survey of multi-interest recommendation research. We first motivate the necessity of multi-interest modeling (Why), then formalize its core paradigms and technical components (What), and finally synthesize reproducible implementation pathways (How)—establishing, for the first time, a unified, problem-driven three-layer analytical framework. We propose a standardized taxonomy covering clustering-, sequence-based, graph neural network-, and generative approaches to multi-interest extraction, and integrate state-of-the-art techniques including attention mechanisms, contrastive learning, and disentangled representation learning. Based on over 120 scholarly works, we construct a comprehensive landscape map and open-source the first unified, multi-paradigm codebase (on GitHub). This work provides both theoretical foundations and practical benchmarks for algorithmic innovation and industrial deployment.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Intelligent Robots: Multi-Robot Systems

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Existing recommendation methods often struggle to model users' multifaceted preferences due to the diversity and volatility of user behavior, as well as the inherent uncertainty and ambiguity of item attributes in practical scenarios. Multi-interest recommendation addresses this challenge by extracting multiple interest representations from users' historical interactions, enabling fine-grained preference modeling and more accurate recommendations. It has drawn broad interest in recommendation research. However, current recommendation surveys have either specialized in frontier recommendation methods or delved into specific tasks and downstream applications. In this work, we systematically review the progress, solutions, challenges, and future directions of multi-interest recommendation by answering the following three questions: (1) Why is multi-interest modeling significantly important for recommendation? (2) What aspects are focused on by multi-interest modeling in recommendation? and (3) How can multi-interest modeling be applied, along with the technical details of the representative modules? We hope that this survey establishes a fundamental framework and delivers a preliminary overview for researchers interested in this field and committed to further exploration. The implementation of multi-interest recommendation summarized in this survey is maintained at https://github.com/WHUIR/Multi-Interest-Recommendation-A-Survey.

Problem

• Modeling diverse and volatile user behavior in recommendations
• Addressing uncertainty in item attributes for accurate recommendations
• Systematically reviewing multi-interest modeling progress and solutions

Innovation

• Extracts multiple interest representations from user history
• Systematically reviews multi-interest recommendation progress
• Provides technical details of representative modules