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

TL;DR

MergeRec introduces a novel model merging framework for data-isolated cross-domain sequential recommendation, enabling better generalization without sharing raw user data, through pseudo-user data construction and collaborative merging optimization.

Contribution

The paper proposes MergeRec, a new approach that effectively merges models across domains without data sharing, addressing privacy constraints and improving recommendation performance.

Findings

MergeRec outperforms traditional methods with up to 17.21% improvement in Recall@10.

The framework successfully enhances model generalization to unseen domains.

Pseudo-user data construction enables meaningful training without real user interactions.

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.