Meta Lattice: Model Space Redesign for Cost-Effective Industry-Scale Ads Recommendations
Abstract
The rapidly evolving landscape of products, surfaces, policies, and regulations poses significant challenges for deploying state-of-the-art recommendation models at industry scale, primarily due to data fragmentation across domains and escalating infrastructure costs that hinder sustained quality improvements. To address this challenge, we propose Lattice, a recommendation framework centered around model space redesign that extends Multi-Domain, Multi-Objective (MDMO) learning beyond models and learning objectives. Lattice addresses these challenges through a comprehensive model space redesign that combines cross-domain knowledge sharing, data consolidation, model unification, distillation, and system optimizations to achieve significant improvements in both quality and cost-efficiency. Our deployment of Lattice at Meta has resulted in 10% revenue-driving top-line metrics gain, 11.5% user satisfaction improvement, 6% boost in conversion rate, with 20% capacity saving.
Cite
@article{arxiv.2512.09200,
title = {Meta Lattice: Model Space Redesign for Cost-Effective Industry-Scale Ads Recommendations},
author = {Liang Luo and Yuxin Chen and Zhengyu Zhang and Mengyue Hang and Andrew Gu and Buyun Zhang and Boyang Liu and Chen Chen and Chengze Fan and Dong Liang and Fan Yang and Feifan Gu and Huayu Li and Jade Nie and Jiayi Xu and Jiyan Yang and Jongsoo Park and Laming Chen and Longhao Jin and Qianru Li and Qin Huang and Shali Jiang and Shiwen Shen and Shuaiwen Wang and Sihan Zeng and Siyang Yuan and Tongyi Tang and Weilin Zhang and Wenjun Wang and Xi Liu and Xiaohan Wei and Xiaozhen Xia and Yuchen Hao and Yunlong He and Yasmine Badr and Zeliang Chen and Maxim Naumov and Yantao Yao and Wenlin Chen and Santanu Kolay and GP Musumeci and Ellie Dingqiao Wen},
journal= {arXiv preprint arXiv:2512.09200},
year = {2025}
}
Comments
Accepted to KDD 2026