NEAR$^2$: A Nested Embedding Approach to Efficient Product Retrieval and Ranking

2025-06-24

AI Summary

E-commerce information retrieval faces dual challenges: low accuracy in user intent understanding and poor efficiency when searching over large-scale product catalogs. To address these, we propose a nested embedding method that preserves the Transformer encoder architecture while introducing a hierarchical embedding compression mechanism. During inference, this method significantly reduces embedding dimensionality—thereby improving computational efficiency—without compromising accuracy or increasing training cost. Our approach jointly optimizes retrieval and ranking via a hybrid objective combining multi-negative ranking loss and online contrastive loss. Experiments on multiple benchmark datasets demonstrate up to 12× improvement in embedding efficiency. Notably, the method achieves superior performance in challenging scenarios such as short queries and implicit user intent. Overall, it establishes a scalable new paradigm for high-accuracy, low-latency e-commerce search.

Technology Category

• Search and Optimization: Learning to Search
• Data Mining & Knowledge Management: Data Compression
• Machine Learning: Learning Preferences or Rankings

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

E-commerce information retrieval (IR) systems struggle to simultaneously achieve high accuracy in interpreting complex user queries and maintain efficient processing of vast product catalogs. The dual challenge lies in precisely matching user intent with relevant products while managing the computational demands of real-time search across massive inventories. In this paper, we propose a Nested Embedding Approach to product Retrieval and Ranking, called NEAR$^2$, which can achieve up to $12$ times efficiency in embedding size at inference time while introducing no extra cost in training and improving performance in accuracy for various encoder-based Transformer models. We validate our approach using different loss functions for the retrieval and ranking task, including multiple negative ranking loss and online contrastive loss, on four different test sets with various IR challenges such as short and implicit queries. Our approach achieves an improved performance over a smaller embedding dimension, compared to any existing models.

Problem

• Improving accuracy and efficiency in e-commerce product retrieval
• Matching user intent with relevant products in real-time
• Reducing embedding size without sacrificing retrieval performance

Innovation

• Nested embedding for efficient product retrieval
• Reduced embedding size with no training cost
• Improved accuracy in Transformer-based models