Customized Retrieval-Augmented Generation with LLM for Debiasing Recommendation Unlearning

TL;DR

This paper introduces CRAGRU, a retrieval-augmented generation framework utilizing large language models to efficiently and effectively unlearn user data in recommender systems, reducing bias and preserving recommendation quality.

Contribution

CRAGRU is a novel RAG-based approach that decouples retrieval and generation for user-specific unlearning, improving efficiency and bias mitigation in recommender systems.

Findings

CRAGRU effectively unlearns user data with minimal impact on non-target users.

It maintains recommendation performance comparable to fully retrained models.

Experiments on three datasets validate its efficiency and bias mitigation capabilities.

Abstract

Modern recommender systems face a critical challenge in complying with privacy regulations like the 'right to be forgotten': removing a user's data without disrupting recommendations for others. Traditional unlearning methods address this by partial model updates, but introduce propagation bias--where unlearning one user's data distorts recommendations for behaviorally similar users, degrading system accuracy. While retraining eliminates bias, it is computationally prohibitive for large-scale systems. To address this challenge, we propose CRAGRU, a novel framework leveraging Retrieval-Augmented Generation (RAG) for efficient, user-specific unlearning that mitigates bias while preserving recommendation quality. CRAGRU decouples unlearning into distinct retrieval and generation stages. In retrieval, we employ three tailored strategies designed to precisely isolate the target user's data influence, minimizing collateral impact on unrelated users and enhancing unlearning efficiency. Subsequently, the generation stage utilizes an LLM, augmented with user profiles integrated into prompts, to reconstruct accurate and personalized recommendations without needing to retrain the entire base model. Experiments on three public datasets demonstrate that CRAGRU effectively unlearns targeted user data, significantly mitigating unlearning bias by preventing adverse impacts on non-target users, while maintaining recommendation performance comparable to fully trained original models. Our work highlights the promise of RAG-based architectures for building robust and privacy-preserving recommender systems. The source code is available at: https://github.com/zhanghaichao520/LLM_rec_unlearning.