Implementing Large Language Models to Enhance Catalog Accuracy in Retail

Abstract

The retail industry faces significant challenges in maintaining accurate product catalogs, which are essential for customer satisfaction, operational efficiency, and data-driven decision-making. Inaccurate or inconsistent product listings can lead to poor user experiences, inventory mismanagement, and lost sales. This paper explores the application of Large Language Models (LLMs) to enhance catalog accuracy in retail environments. By leveraging advanced natural language processing techniques, LLMs can automate the extraction and categorization of product information from unstructured data sources, such as supplier descriptions and customer reviews. Additionally, LLMs can assist in detecting discrepancies, suggesting improvements, and ensuring consistent language use across catalog entries. The study demonstrates how integrating LLMs into catalog management systems not only improves the quality and accuracy of product listings but also reduces manual effort and operational costs. Furthermore, the paper discusses the potential for LLMs to scale as product catalogs grow, enabling continuous enhancement of retail catalog accuracy in a dynamic market.