RecGPT-Mobile: On-Device Large Language Models for User Intent Understanding in Taobao Feed Recommendation

TL;DR

RecGPT-Mobile introduces a lightweight on-device LLM framework for real-time user intent understanding in mobile e-commerce, enhancing recommendation accuracy while reducing inference costs.

Contribution

It presents a novel on-device LLM-based intent understanding agent tailored for mobile e-commerce, enabling faster adaptation to user interests and real-time recommendations.

Findings

Significant improvement in recommendation accuracy demonstrated through experiments.

On-device deployment reduces inference costs compared to cloud-based solutions.

Framework enables real-time user interest capture and adjustment.

Abstract

Predicting a user's next search query from recent interaction behaviors is a critical problem in modern e-commerce systems, particularly in scenarios where user intent evolves rapidly. Large Language Models (LLMs) offer strong semantic reasoning capabilities and have recently been adopted to enhance training data construction for next-query prediction. However, due to resource constraints on mobile devices, existing applications are deployed on cloud servers, resulting in high inference costs. In this paper, we propose RecGPT-Mobile, a framework that designs a lightweight LLM-based intent understanding agent to improve recommendation quality in mobile e-commerce scenarios. By deploying LLMs directly on mobile devices, our approach can capture evolving interests of users more quickly and adjust the recommendation results in real time. Extensive offline analyses and online experiments demonstrate that our method significantly improves the accuracy of recommendation results, laying a practical path for LLM deployment in production-scale recommendation systems on mobile devices, as well as a scalable solution for integrating LLMs into real-world next-query prediction systems.