Attribute-Aware Controlled Product Generation with LLMs for E-commerce

TL;DR

This paper introduces a systematic method using Large Language Models to generate high-quality synthetic e-commerce product data with controlled attributes, enhancing dataset quality for improved product information extraction.

Contribution

The paper presents a novel attribute-aware controlled data generation framework using LLMs, enabling high-quality synthetic product data creation for e-commerce applications.

Findings

Synthetic data achieves 60.5% accuracy on MAVE dataset, comparable to real data at 60.8%.

Human evaluation shows 99.6% naturalness and 96.5% valid attributes in synthetic products.

Hybrid synthetic and real data improve accuracy to 68.8%.

Abstract

Product information extraction is crucial for e-commerce services, but obtaining high-quality labeled datasets remains challenging. We present a systematic approach for generating synthetic e-commerce product data using Large Language Models (LLMs), introducing a controlled modification framework with three strategies: attribute-preserving modification, controlled negative example generation, and systematic attribute removal. Using a state-of-the-art LLM with attribute-aware prompts, we enforce store constraints while maintaining product coherence. Human evaluation of 2000 synthetic products demonstrates high effectiveness, with 99.6% rated as natural, 96.5% containing valid attribute values, and over 90% showing consistent attribute usage. On the public MAVE dataset, our synthetic data achieves 60.5% accuracy, performing on par with real training data (60.8%) and significantly improving upon the 13.4% zero-shot baseline. Hybrid configurations combining synthetic and real data further improve performance, reaching 68.8% accuracy. Our framework provides a practical solution for augmenting e-commerce datasets, particularly valuable for low-resource scenarios.