Can I read A Hybrid Cross-Stage Coordination Pre-ranking Model for Online Recommendation Systems on EtoBox?

A Hybrid Cross-Stage Coordination Pre-ranking Model for Online Recommendation Systems by Zhao, Binglei; Qi, Houying; Xu, Guang; Ma, Mian; Zhao, Xiwei; Mei, Feng; Xu, Sulong; Hu, Jinghe is a scholarly article available to read on EtoBox.

What is A Hybrid Cross-Stage Coordination Pre-ranking Model for Online Recommendation Systems about?

Large-scale recommendation systems often adopt cascading architecture consisting of retrieval, pre-ranking, ranking, and re-ranking stages. With strict latency requirements, pre-ranking utilizes lightweight models to perform a preliminary selection from massive retrieved candidates. However, recent works focus solely on improving consistency with ranking, relying exclusively on downstream stages. Since downstream input is derived from the pre-ranking output, they will exacerbate the sample selection bias (SSB) issue and Matthew effect, leading to sub-optimal results. To address the limitation, we propose a novel Hybrid Cross-Stage Coordination Pre-ranking model (HCCP) to integrate information from upstream (retrieval) and downstream (ranking, re-ranking) stages. Specifically, cross-stage coordination refers to the pre-ranking's adaptability to the entire stream and the role of serving as a more effective bridge between upstream and downstream. HCCP consists of Hybrid Sample Construction and Hybrid Objective Optimization. Hybrid sample construction captures multi-level unexposed data from the entire stream and rearranges them to become the optimal guiding "ground truth" for pre-rank

Author

Zhao, Binglei; Qi, Houying; Xu, Guang; Ma, Mian; Zhao, Xiwei; Mei, Feng; Xu, Sulong; Hu, Jinghe

Published

2025

Language

EN