RecGPT Technical Report

2025-07-30

AI Summary

Industrial recommender systems overly rely on historical co-occurrence patterns and log fitting, neglecting explicit user intent modeling—leading to interest overfitting, filter bubbles, and long-tail entrenchment. To address this, we propose RecGPT, the first intent-centric recommendation framework that unifies intent discovery, multi-granularity retrieval, and interpretable generation via large language models (LLMs). We design a two-stage training mechanism: reasoning-enhanced pre-alignment followed by self-training evolution. Furthermore, we introduce a human-in-the-loop evaluation system to guide iterative optimization. Deployed at scale in the Taobao mobile application, online A/B tests demonstrate simultaneous improvements in user satisfaction, content diversity, merchant exposure, and platform conversion rate. These results empirically validate the critical role of intent-driven recommendation in fostering sustainable ecosystem health.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Natural Language Processing: Generation

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Recommender systems are among the most impactful applications of artificial intelligence, serving as critical infrastructure connecting users, merchants, and platforms. However, most current industrial systems remain heavily reliant on historical co-occurrence patterns and log-fitting objectives, i.e., optimizing for past user interactions without explicitly modeling user intent. This log-fitting approach often leads to overfitting to narrow historical preferences, failing to capture users' evolving and latent interests. As a result, it reinforces filter bubbles and long-tail phenomena, ultimately harming user experience and threatening the sustainability of the whole recommendation ecosystem. To address these challenges, we rethink the overall design paradigm of recommender systems and propose RecGPT, a next-generation framework that places user intent at the center of the recommendation pipeline. By integrating large language models (LLMs) into key stages of user interest mining, item retrieval, and explanation generation, RecGPT transforms log-fitting recommendation into an intent-centric process. To effectively align general-purpose LLMs to the above domain-specific recommendation tasks at scale, RecGPT incorporates a multi-stage training paradigm, which integrates reasoning-enhanced pre-alignment and self-training evolution, guided by a Human-LLM cooperative judge system. Currently, RecGPT has been fully deployed on the Taobao App. Online experiments demonstrate that RecGPT achieves consistent performance gains across stakeholders: users benefit from increased content diversity and satisfaction, merchants and the platform gain greater exposure and conversions. These comprehensive improvement results across all stakeholders validates that LLM-driven, intent-centric design can foster a more sustainable and mutually beneficial recommendation ecosystem.

Problem

• Overcoming reliance on historical co-occurrence patterns in recommender systems
• Addressing overfitting to narrow user preferences and filter bubbles
• Enhancing user intent modeling for sustainable recommendation ecosystems

Innovation

• Integrates LLMs for intent-centric recommendations
• Uses multi-stage training with reasoning-enhanced alignment
• Deploys Human-LLM cooperative judge system