PriCoRec: A Privacy-Aware Cloud-Device Collaborative Framework for Ad Recommendation under Feature Constraints

2026-08-14

AI Summary

This study addresses the challenge of balancing personalization with regulatory compliance in cloud-based recommendation systems. We propose PriCoRec, a privacy-preserving cloud-device collaborative advertising recommendation framework. Adopting a two-stage architecture comprising cloud-based pre-ranking and on-device re-ranking, this method retains sensitive data locally through feature decoupling. Furthermore, it introduces diversity regularization to optimize candidate quality and employs a cloud-guided lightweight training mechanism to balance performance with power consumption. Experimental results demonstrate that PriCoRec maintains robust recommendation accuracy while ensuring sensitive features never leave the device. Simultaneously, it effectively controls on-device computational overhead and energy consumption, successfully achieving a unified solution for privacy protection and efficient personalization under strict regulatory constraints.

Abstract

Privacy regulations increasingly restrict cloud processing of sensitive user data (e.g., age, gender), hindering traditional cloud-only recommendation models. To mitigate this challenge, we propose a Privacy-aware Collaborative cloud-device ads Recommendation framework (PriCoRec) which personalizes recommendations while keeping sensitive features on-device. While separating recommendation into cloud-based and on-device stages enables privacy-aware deployment, naive splitting suffers from degraded shortlist quality and inefficient on-device inference due to limited private features. We therefore design a collaborative framework that comprises a cloud-based pre-ranking stage using cloud-accessible features, and an on-device ranking stage that locally incorporates highly personalized features. We introduce a diversity regularizer to pre-ranking to improve candidate quality. Moreover, to control device power consumption and computational cost, we incorporate a cloud-guided training mechanism that enhances device model performance while keeping the model lightweight. Experiments demonstrate that the proposed framework maintains strong recommendation performance while keeping sensitive features on-device.

Problem

• Privacy-aware Recommendation
• Cloud-Device Collaboration
• Feature Constraints
• Ad Recommendation

Innovation

• Privacy-Aware Cloud-Device Collaboration
• Diversity Regularizer
• Cloud-Guided Training
• On-Device Ranking
• Lightweight Model