FedCGR: Federated Cross-Domain Generative Recommendation

2026-08-11

AI Summary

This work addresses the challenges in federated cross-domain recommendation, where sparse user overlap, unavailability of interaction data, and privacy constraints hinder effective cross-domain item alignment. To overcome these issues, the authors propose a generative federated recommendation framework grounded in semantic item language. By leveraging public item metadata to construct discrete semantic ID sequences, the method enables cross-domain alignment without exchanging private interaction data. A reliability-aware semantic interface is designed to inject local collaborative filtering signals, and a prototype-based personalized generator selectively aggregates shared parameters according to domain relevance while preserving domain-specific characteristics. Evaluated across six Amazon cross-domain scenarios, the proposed approach significantly outperforms existing federated generative baselines and achieves performance on par with state-of-the-art sequential and federated cross-domain recommendation methods under both full-ranking and sampled evaluation protocols.

Abstract

Cross-domain recommendation (CDR) transfers preference knowledge across related domains, but federated deployment makes cross-domain alignment difficult because the behavioral anchors that align item spaces, such as overlapping users and shared interaction signals, are often sparse, unavailable, or privacy-sensitive across clients. To address this tension, we revisit federated CDR as generation over a stable semantic item language. By representing items as discrete semantic ID (SID) sequences derived from public item-side metadata, cross-domain item alignment is induced by a shared vocabulary rather than by exchanging private interactions or aligning domain-specific embeddings. Directly federating SID-based generators, however, introduces two design constraints: the SID tokenizer must remain fixed to preserve cross-client token consistency, which creates a semantic-only bottleneck because local collaborative filtering (CF) signals cannot be globally shared or aligned; meanwhile, standard federated averaging can cause negative transfer under domain heterogeneity. To overcome these constraints, we propose FedCGR, a federated generative CDR framework that keeps the item language stable and makes adaptation explicit. FedCGR injects local CF evidence through a reliability-aware semantic interface and trains a prototype-personalized generator that selectively aggregates shared parameters according to domain relatedness while keeping domain-specific quantities local. Experiments on six Amazon cross-domain scenarios show that FedCGR consistently outperforms federated generative baselines and achieves competitive performance against strong sequential and federated CDR methods under both full-ranking and sampled evaluation protocols.

Problem

• federated learning
• cross-domain recommendation
• item alignment
• privacy
• semantic representation

Innovation

• Federated Learning
• Cross-Domain Recommendation
• Generative Recommendation
• Semantic Item Representation
• Privacy-Preserving AI