Cross-Country Code-Mixing for Generative Recommendation

2026-09-23

AI Summary

This study addresses the absence of data-level knowledge transfer caused by ID isolation in cross-national e-commerce recommendation, proposing the CMRec framework. Inspired by code-switching in natural language processing, this work introduces a novel dual-constrained, context-aware code-mixing mechanism. By integrating multimodal semantic codebook learning, token-level sequence substitution, and a context-aware reweighted loss function, it generates cross-national mixed behavior sequences to achieve deep knowledge transfer and unified modeling from the parameter level to the data level. Experimental results demonstrate that the proposed method significantly improves recommendation quality for data-sparse countries. Furthermore, A/B testing on a large-scale e-commerce platform yields a 1.77% increase in advertising revenue and a 2.64% growth in order volume.

Abstract

Cross-country recommendation on modern e-commerce platforms is typically deployed with disjoint user and item ID spaces across markets, removing the shared anchors that conventional cross-domain methods rely on. Generative recommendation (GR) mitigates this by mapping items into a shared token space and training a unified model, but existing approaches keep behavior sequences strictly country-specific, so knowledge transfer occurs only at the parameter level and remains absent at the data level. Inspired by code-switching corpora in multilingual natural language processing, we propose CMRec, a cross-country GR framework that injects cross-country supervision at the data level via dual-constrained, context-aware code-mixing. CMRec first learns a shared semantic codebook from multi-modal content and behavioral co-occurrence across countries. It then uses this codebook to synthesize mixed-country sequences via token-level substitutions that satisfy both static (content) and dynamic (e.g., price, audience, popularity) constraints. Finally, it introduces a context-aware loss that reweights mixed samples according to their plausibility in the current sequence. Experiments on two real-world multi-country datasets and an online A/B test show that CMRec substantially improves recommendation quality in data-sparse countries while preserving performance in data-rich countries, achieving +1.77% advertising revenue and +2.64% orders on a large-scale e-commerce platform.

Problem

• Cross-Country Recommendation
• Generative Recommendation
• Code-Mixing
• Disjoint ID Spaces
• Knowledge Transfer

Innovation

• Generative Recommendation
• Code-Mixing
• Cross-Country
• Shared Semantic Codebook
• Context-Aware Loss