Mosaic: A Fleet of User Embedding Specialists for Recommendation at Meta
2026-07-27 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors developed Mosaic, a new system that combines different types of user models to better understand user behavior in recommendation systems. Each model focuses on different user activity patterns, and special training methods help these models contribute unique information. They also created ways to evaluate these user embeddings without extra data logging, making development faster. Mosaic runs efficiently on both CPUs and GPUs and shows improvements in both offline tests and online performance.
user representationembedding vectorsrecommendation systemsmulti-task learningembedding evaluationCPU/GPU servinguser behavior modelingMosaic platformCosine Redundancy LossMulti-task Relations Mining
Authors
John Zhiyuan Zheng, Xian Sun, Xiangyang Mou, Yujunrong Ma, Christina You, Michael Jiayuan He, Hrishikesh Paranjape, Aakarsha Agarwal, Hong Li
Abstract
User representation is one of the highest-leverage modeling problems in industrial recommendation systems: a single advancement in how users are encoded can propagate across retrieval, ranking, and integrity tasks at platform scale. Prior industrial user representation work builds either a single user model that emits one or more embedding vectors or a shared backbone with task-specific adaptation. In this paper, we present Mosaic, a foundational user modeling platform that employs a fleet of specialists to learn user embeddings. The fleet comprises four architecturally diverse model families - memorization-driven, dense-heavy, sequential-based, and CoTrain models - each focusing on a distinct facet of user behavior. We developed MRM (Multi-task Relations Mining) and CRL (Cosine Redundancy Loss) techniques to maximize the marginal information contribution of each new specialist. We also introduce CoEval and User Tower Zero-Out, new logging-free embedding evaluation framework that improves development velocity while preserving downstream-aligned accuracy. Our hybrid CPU/GPU, online-and-offline serving stack allows each specialist to choose the adequate serving strategy to meet the freshness, latency, and computational requirements. Mosaic delivers consistent and significant offline NE improvements in addition to online gains.