MBGR: Multi-Business Prediction for Generative Recommendation at Meituan

2026-04-02

AI Summary

This work addresses the trade-off dilemma and representation ambiguity caused by unified semantic IDs in generative recommendation across multiple business scenarios. To this end, we propose MBGR, a multi-business generative recommendation framework that introduces three key components: business-aware semantic IDs (BIDs), a multi-business prediction (MBP) architecture, and a label-based dynamic routing (LDR) mechanism. Operating under the Next Token Prediction paradigm, MBGR enables effective cross-business joint modeling while preserving business-specific semantics, thereby mitigating interference and semantic confusion. As the first generative recommendation system tailored for multi-business settings, MBGR demonstrates significant performance gains in both offline and online experiments on Meituan’s food delivery platform, and has been successfully deployed in production, validating its industrial-scale effectiveness.

Abstract

Generative recommendation (GR) has recently emerged as a promising paradigm for industrial recommendations. GR leverages Semantic IDs (SIDs) to reduce the encoding-decoding space and employs the Next Token Prediction (NTP) framework to explore scaling laws. However, existing GR methods suffer from two critical issues: (1) a \textbf{seesaw phenomenon} in multi-business scenarios arises due to NTP's inability to capture complex cross-business behavioral patterns; and (2) a unified SID space causes \textbf{representation confusion} by failing to distinguish distinct semantic information across businesses. To address these issues, we propose Multi-Business Generative Recommendation (MBGR), the first GR framework tailored for multi-business scenarios. Our framework comprises three key components. First, we design a Business-aware semantic ID (BID) module that preserves semantic integrity via domain-aware tokenization. Then, we introduce a Multi-Business Prediction (MBP) structure to provide business-specific prediction capabilities. Furthermore, we develop a Label Dynamic Routing (LDR) module that transforms sparse multi-business labels into dense labels to further enhance the multi-business generation capability. Extensive offline and online experiments on Meituan's food delivery platform validate MBGR's effectiveness, and we have successfully deployed it in production.

Problem

• Generative Recommendation
• Multi-Business Scenarios
• Seesaw Phenomenon
• Representation Confusion
• Semantic ID

Innovation

• Generative Recommendation
• Multi-Business Prediction
• Semantic ID
• Label Dynamic Routing
• Business-aware Tokenization