Personalized Transformer-based Ranking for e-Commerce at Yandex

TL;DR

This paper presents a transformer-based two-tower model for personalized e-commerce recommendations, emphasizing the importance of the ranking stage, and demonstrates significant offline and online performance improvements at Yandex.

Contribution

The paper introduces a novel two-stage training process and a transformer-based architecture specifically designed for personalized ranking in e-commerce.

Findings

Significant offline ranking improvements using the proposed model.

Successful deployment serving millions of users daily.

Strong online A/B testing results confirming effectiveness.

Abstract

Personalizing user experience with high-quality recommendations based on user activity is vital for e-commerce platforms. This is particularly important in scenarios where the user's intent is not explicit, such as on the homepage. Recently, personalized embedding-based systems have significantly improved the quality of recommendations and search in the e-commerce domain. However, most of these works focus on enhancing the retrieval stage. In this paper, we demonstrate that features produced by retrieval-focused deep learning models are sub-optimal for ranking stage in e-commerce recommendations. To address this issue, we propose a two-stage training process that fine-tunes two-tower models to achieve optimal ranking performance. We provide a detailed description of our transformer-based two-tower model architecture, which is specifically designed for personalization in e-commerce. Additionally, we introduce a novel technique for debiasing context in offline models and report significant improvements in ranking performance when using web-search queries for e-commerce recommendations. Our model has been successfully deployed at Yandex, serves millions of users daily, and has delivered strong performance in online A/B testing.