DOS: Dual-Flow Orthogonal Semantic IDs for Recommendation in Meituan

TL;DR

This paper introduces DOS, a dual-flow orthogonal semantic ID method that improves recommendation quality by aligning semantic codebooks with generation space and preserving semantics, demonstrated through extensive experiments and deployment.

Contribution

The paper proposes a novel dual-flow framework with orthogonal residual quantization to enhance semantic preservation and alignment in recommendation systems.

Findings

DOS outperforms existing methods in offline and online tests.

Successfully deployed in Meituan's app serving hundreds of millions of users.

Significantly improves recommendation accuracy and semantic fidelity.

Abstract

Semantic IDs serve as a key component in generative recommendation systems. They not only incorporate open-world knowledge from large language models (LLMs) but also compress the semantic space to reduce generation difficulty. However, existing methods suffer from two major limitations: (1) the lack of contextual awareness in generation tasks leads to a gap between the Semantic ID codebook space and the generation space, resulting in suboptimal recommendations; and (2) suboptimal quantization methods exacerbate semantic loss in LLMs. To address these issues, we propose Dual-Flow Orthogonal Semantic IDs (DOS) method. Specifically, DOS employs a user-item dual flow-framework that leverages collaborative signals to align the Semantic ID codebook space with the generation space. Furthermore, we introduce an orthogonal residual quantization scheme that rotates the semantic space to an appropriate orientation, thereby maximizing semantic preservation. Extensive offline experiments and online A/B testing demonstrate the effectiveness of DOS. The proposed method has been successfully deployed in Meituan's mobile application, serving hundreds of millions of users.