Retrieval-Augmented Purifier for Robust LLM-Empowered Recommendation
ACM Transactions on Information Systems · 31.05.2026 · 44 (4) · 1-42
DOI: 10.1145/3795521 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Information Systems, online erschienen am 16.04.2026. Verfasst von Liangbo Ning und 2 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 96 Literaturangaben. Gefördert von: General Research Funds from the Hong Kong Research Grants Council, Hong Kong Research Grants Council’s Theme-based Research Scheme, Hong Kong Research Grants Council’s Research Impact Fund.
Abstract (Originalsprache)
Recently, Large Language Model (LLM)-empowered recommender systems have revolutionized personalized recommendation frameworks and attracted extensive attention. Despite the remarkable success, existing LLM-empowered RecSys have been demonstrated to be highly vulnerable to minor perturbations. To mitigate the negative impact of such vulnerabilities, one potential solution is to employ collaborative signals based on item–item co-occurrence to purify the malicious collaborative knowledge from the user’s irrelevant historical interactions. On the other hand, due to the capabilities to expand insufficient internal knowledge of LLMs, Retrieval-Augmented Generation (RAG) techniques provide unprecedented opportunities to enhance the robustness of LLM-empowered recommender systems by introducing external collaborative knowledge. Therefore, in this article, we propose a novel framework ( RETURN ) by retrieving external collaborative signals to purify the poisoned user profiles and enhance the robustness of LLM-empowered RecSys in a plug-and-play manner. Specifically, retrieval-augmented perturbation positioning is proposed to identify potential perturbations within the users’ historical sequences by retrieving external knowledge from collaborative item graphs. After that, we further retrieve the collaborative knowledge to cleanse the perturbations by using either deletion or replacement strategies and introduce a robust ensemble recommendation strategy to generate final robust predictions. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed RETURN.
Förderung
- General Research Funds from the Hong Kong Research Grants Council
- Hong Kong Research Grants Council’s Theme-based Research Scheme
- Hong Kong Research Grants Council’s Research Impact Fund
- Hong Kong Research Grants Council’s Collaborative Research Fund
- internal research funds from Hong Kong Polytechnic University
Zitieren
APA
Ning, L., Fan, W., & Li, Q. (2026). Retrieval-Augmented Purifier for Robust LLM-Empowered Recommendation. ACM Transactions on Information Systems, 44(4), 1-42. https://doi.org/10.1145/3795521
Vancouver
Ning L, Fan W, Li Q. Retrieval-Augmented Purifier for Robust LLM-Empowered Recommendation. ACM Transactions on Information Systems. 2026;44(4):1-42. doi:10.1145/3795521
BibTeX
@article{Ning2026,
author = {Ning, Liangbo and Fan, Wenqi and Li, Qing},
title = {Retrieval-Augmented Purifier for Robust LLM-Empowered Recommendation},
journal = {ACM Transactions on Information Systems},
year = {2026},
volume = {44},
number = {4},
pages = {1--42},
doi = {10.1145/3795521}
}
EUVIMED