IPL: Leveraging Multimodal Large Language Models for Intelligent Product Listing

TL;DR

This paper introduces IPL, an AI-powered tool that uses multimodal large language models to help individual sellers create product descriptions from photos, improving listing quality on C2C platforms.

Contribution

It develops a domain-specific multimodal AI system for automatic product description generation tailored to C2C e-commerce platforms.

Findings

IPL outperforms base models in domain-specific tasks.

Generated descriptions have fewer hallucinations.

User engagement increased with AI-assisted listings.

Abstract

Unlike professional Business-to-Consumer (B2C) e-commerce platforms (e.g., Amazon), Consumer-to-Consumer (C2C) platforms (e.g., Facebook marketplace) are mainly targeting individual sellers who usually lack sufficient experience in e-commerce. Individual sellers often struggle to compose proper descriptions for selling products. With the recent advancement of Multimodal Large Language Models (MLLMs), we attempt to integrate such state-of-the-art generative AI technologies into the product listing process. To this end, we develop IPL, an Intelligent Product Listing tool tailored to generate descriptions using various product attributes such as category, brand, color, condition, etc. IPL enables users to compose product descriptions by merely uploading photos of the selling product. More importantly, it can imitate the content style of our C2C platform Xianyu. This is achieved by employing domain-specific instruction tuning on MLLMs and adopting the multi-modal Retrieval-Augmented Generation (RAG) process. A comprehensive empirical evaluation demonstrates that the underlying model of IPL significantly outperforms the base model in domain-specific tasks while producing less hallucination. IPL has been successfully deployed in our production system, where 72% of users have their published product listings based on the generated content, and those product listings are shown to have a quality score 5.6% higher than those without AI assistance.