PulseAugur
EN
LIVE 21:57:51

Alibaba deploys MGOE for billion-scale recommendations, boosting performance

By PulseAugur Editorial · [1 sources] ·

Researchers have developed the Macro Graph of Experts (MGOE) framework, a novel approach for multi-task recommendation systems operating at a billion-scale. This framework uniquely incorporates graph information by leveraging macro graph embeddings to capture task-specific features and model correlations between experts. MGOE has been successfully deployed by Alibaba for its billion-scale recommender system, demonstrating significant improvements in both offline experiments and online A/B tests compared to existing state-of-the-art methods. AI

IMPACT Enhances large-scale recommendation systems by incorporating graph structures for improved performance and task correlation modeling.

RANK_REASON Publication of a research paper detailing a new framework for large-scale recommendation systems. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Alibaba deploys MGOE for billion-scale recommendations, boosting performance

COVERAGE [1]

  1. arXiv cs.LG TIER_1 English(EN) · Hongyu Yao, Zijin Hong, Hao Chen, Zhiqing Li, Qijie Shen, Zuobin Ying, Qihua Feng, Huan Gong, Feiran Huang ·

    Macro Graph of Experts for Billion-Scale Multi-Task Recommendation

    arXiv:2506.10520v5 Announce Type: replace-cross Abstract: Graph-based multi-task learning at billion-scale presents a significant challenge, as different tasks correspond to distinct billion-scale graphs. Traditional multi-task learning methods often neglect these graph structure…

RELATED ENTITIES

RELATED TOPICS