Enhancing Graph-based Recommendations with Majority-Voting LLM-Rerank Augmentation
Machine Learning · 2026-05 · 115 (5)
DOI: 10.1007/s10994-026-07051-9 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning, online erschienen am 09.05.2026. Verfasst von Minh-Anh Nguyen und 5 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Gefördert von: VinUni Center for AI Research.
Abstract (Originalsprache)
Recommendation systems often suffer from data sparsity, caused by limited user-item interactions, which degrades their performance and amplifies popularity bias in real-world scenarios. This paper proposes a novel data augmentation framework that leverages Large Language Models (LLMs) and item textual descriptions to enrich interaction data. By few-shot prompting LLMs multiple times to rerank items and aggregating the results via majority voting, we generate high-confidence synthetic user-item interactions, supported by theoretical guarantees based on the concentration of measure. To effectively leverage the augmented data in the context of a graph recommendation system, we integrate it into a graph contrastive learning framework to mitigate distributional shift and alleviate popularity bias. Extensive experiments show that our method improves accuracy and reduces popularity bias, outperforming strong baselines.
Förderung
- VinUni Center for AI Research
Zitieren
APA
Nguyen, M., Nguyen, B., Lan, N. T. H., Hoang, T., Le, D., & Le, D. D. (2026). Enhancing Graph-based Recommendations with Majority-Voting LLM-Rerank Augmentation. Machine Learning, 115(5). https://doi.org/10.1007/s10994-026-07051-9
Vancouver
Nguyen M, Nguyen B, Lan NTH, Hoang T, Le D, Le DD. Enhancing Graph-based Recommendations with Majority-Voting LLM-Rerank Augmentation. Machine Learning. 2026;115(5). doi:10.1007/s10994-026-07051-9
BibTeX
@article{Nguyen2026,
author = {Nguyen, Minh-Anh and Nguyen, Bao and Lan, N. T. Ha and Hoang, Tuan-Anh and Le, Duc-Trong and Le, Dung D.},
title = {Enhancing Graph-based Recommendations with Majority-Voting LLM-Rerank Augmentation},
journal = {Machine Learning},
year = {2026},
volume = {115},
number = {5},
doi = {10.1007/s10994-026-07051-9}
}
EUVIMED