Semantic Retrieval at Walmart

TL;DR

This paper introduces a hybrid neural and traditional retrieval system for Walmart's e-commerce search, significantly enhancing relevance for tail queries through scalable training and practical deployment strategies.

Contribution

It presents a novel hybrid retrieval system combining inverted index and neural embeddings, with a scalable training technique and deployment insights for large-scale e-commerce search.

Findings

Significant improvement in search relevance for tail queries

Effective deployment with minimal response time impact

Successful integration of neural models in production environment

Abstract

In product search, the retrieval of candidate products before re-ranking is more critical and challenging than other search like web search, especially for tail queries, which have a complex and specific search intent. In this paper, we present a hybrid system for e-commerce search deployed at Walmart that combines traditional inverted index and embedding-based neural retrieval to better answer user tail queries. Our system significantly improved the relevance of the search engine, measured by both offline and online evaluations. The improvements were achieved through a combination of different approaches. We present a new technique to train the neural model at scale. and describe how the system was deployed in production with little impact on response time. We highlight multiple learnings and practical tricks that were used in the deployment of this system.