Leveraging Multimodal Data and Side Users for Diffusion Cross-Domain Recommendation

2025-07-05

AI Summary

To address two key challenges in cross-domain recommendation (CDR)—inadequate cold-start user modeling and insufficient learning of target-domain distribution—this paper proposes a diffusion-based multimodal cross-domain feature generation framework. Methodologically: (1) a multimodal large language model extracts rich, semantic-rich item features across modalities; (2) prompt learning dynamically models users’ cross-domain preferences; (3) auxiliary “side” users—interacting exclusively with the target domain—are introduced to collaboratively refine the target-domain embedding space distribution; and (4) a cross-domain alignment-guided diffusion mechanism generates high-fidelity user and item representations for the target domain. Extensive experiments on the Amazon multi-domain benchmark demonstrate that our approach significantly outperforms state-of-the-art baselines, validating its superiority in both expressive feature representation and accurate distribution modeling.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• 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
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Cross-domain recommendation (CDR) aims to address the persistent cold-start problem in Recommender Systems. Current CDR research concentrates on transferring cold-start users' information from the auxiliary domain to the target domain. However, these systems face two main issues: the underutilization of multimodal data, which hinders effective cross-domain alignment, and the neglect of side users who interact solely within the target domain, leading to inadequate learning of the target domain's vector space distribution. To address these issues, we propose a model leveraging Multimodal data and Side users for diffusion Cross-domain recommendation (MuSiC). We first employ a multimodal large language model to extract item multimodal features and leverage a large language model to uncover user features using prompt learning without fine-tuning. Secondly, we propose the cross-domain diffusion module to learn the generation of feature vectors in the target domain. This approach involves learning feature distribution from side users and understanding the patterns in cross-domain transformation through overlapping users. Subsequently, the trained diffusion module is used to generate feature vectors for cold-start users in the target domain, enabling the completion of cross-domain recommendation tasks. Finally, our experimental evaluation of the Amazon dataset confirms that MuSiC achieves state-of-the-art performance, significantly outperforming all selected baselines. Our code is available: https://anonymous.4open.science/r/MuSiC-310A/.

Problem

• Underutilization of multimodal data in cross-domain recommendations
• Neglect of side users in target domain learning
• Cold-start user feature generation in target domain

Innovation

• Uses multimodal LLM for item feature extraction
• Employs cross-domain diffusion for vector generation
• Leverages side users to enhance target domain learning