Score: 0

Enhancing Graph-based Recommendations with Majority-Voting LLM-Rerank Augmentation

Published: July 29, 2025 | arXiv ID: 2507.21563v1

By: Minh-Anh Nguyen , Bao Nguyen , Ha Lan N. T. and more

Potential Business Impact:

Makes movie suggestions better by adding fake user data.

Business Areas:
Social News Media and Entertainment

Recommendation systems often suffer from data sparsity caused by limited user-item interactions, which degrade their performance and amplify 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.

Page Count
9 pages

Category
Computer Science:
Information Retrieval