MergeRec: Model Merging for Data-Isolated Cross-Domain Sequential Recommendation

2026-01-05

arXiv.org

AI Summary

This work addresses the limited generalization capability in cross-domain sequential recommendation caused by strict data isolation, without relying on user or item overlap across domains. The authors propose MergeRec, a novel framework that enables effective cross-domain modeling without sharing raw interaction data. MergeRec employs a three-stage strategy: training-free model fusion for initialization, item-based pseudo-user sequence generation, and joint optimization combining recommendation loss with knowledge distillation. Notably, MergeRec is the first approach to achieve cross-domain sequential recommendation under stringent data isolation constraints. Experimental results demonstrate substantial improvements in generalization performance, with an average 17.21% gain in Recall@10 across multiple domains, outperforming existing model fusion methods.

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: Federated recommendation systems and personalization
• 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

Modern recommender systems trained on domain-specific data often struggle to generalize across multiple domains. Cross-domain sequential recommendation has emerged as a promising research direction to address this challenge; however, existing approaches face fundamental limitations, such as reliance on overlapping users or items across domains, or unrealistic assumptions that ignore privacy constraints. In this work, we propose a new framework, MergeRec, based on model merging under a new and realistic problem setting termed data-isolated cross-domain sequential recommendation, where raw user interaction data cannot be shared across domains. MergeRec consists of three key components: (1) merging initialization, (2) pseudo-user data construction, and (3) collaborative merging optimization. First, we initialize a merged model using training-free merging techniques. Next, we construct pseudo-user data by treating each item as a virtual sequence in each domain, enabling the synthesis of meaningful training samples without relying on real user interactions. Finally, we optimize domain-specific merging weights through a joint objective that combines a recommendation loss, which encourages the merged model to identify relevant items, and a distillation loss, which transfers collaborative filtering signals from the fine-tuned source models. Extensive experiments demonstrate that MergeRec not only preserves the strengths of the original models but also significantly enhances generalizability to unseen domains. Compared to conventional model merging methods, MergeRec consistently achieves superior performance, with average improvements of up to 17.21% in Recall@10, highlighting the potential of model merging as a scalable and effective approach for building universal recommender systems. The source code is available at https://github.com/DIALLab-SKKU/MergeRec.

Problem

• cross-domain sequential recommendation
• data isolation
• privacy constraints
• model merging
• recommendation generalization

Innovation

• model merging
• cross-domain sequential recommendation
• data-isolated
• pseudo-user data
• collaborative optimization