PulseAugur
EN
LIVE 08:20:35

Airbnb deploys JourneyFormer sequence model for search ranking

Airbnb has developed and deployed JourneyFormer, a sequence modeling solution for its search ranking system. The model addresses challenges such as long and complex guest sequences, sparse booking labels, and the need for scalability. Key design decisions involved guest event selection, ID embeddings, and model architecture, alongside strategies for accelerated training and inference. JourneyFormer has been successfully integrated into Airbnb's production environment, demonstrating improvements in offline ranking metrics and significant gains in online business metrics across multiple surfaces. AI

IMPACT This deployment showcases practical applications of sequence modeling in e-commerce, potentially influencing how other platforms optimize user journeys and search.

RANK_REASON The item is a research paper detailing a sequence modeling solution developed and deployed by Airbnb. [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 →

COVERAGE [1]

  1. arXiv cs.LG TIER_1 English(EN) · Daochen Zha, Chun How Tan, Xin Liu, Bin Xu, Han Zhao, Xiaowei Liu, Tracy Yu, Hui Gao, Huiji Gao, Liwei He, Stephanie Moyerman, Sanjeev Katariya ·

    JourneyFormer: Encoding Airbnb Guest Journey with Sequence Modeling

    arXiv:2606.19108v1 Announce Type: new Abstract: Sequence modeling has become increasingly popular in recommendation and ranking algorithms, owing to its capacity to model users' historical behaviors and infer user intentions. Despite its theoretical simplicity, the practical depl…