Hint-Augmented Re-ranking: Efficient Product Search using LLM-Based Query Decomposition

TL;DR

This paper introduces a hint-augmented re-ranking method that leverages large language models to interpret superlative queries in e-commerce, improving search accuracy while maintaining efficiency.

Contribution

It presents a novel framework for extracting structured hints from queries using LLMs and transferring this knowledge to lightweight models for practical deployment.

Findings

Improved MAP by 10.9 points over baselines

Enhanced MRR by 5.9 points in ranking

Demonstrated effective transfer of superlative semantics

Abstract

Search queries with superlatives (e.g., best, most popular) require comparing candidates across multiple dimensions, demanding linguistic understanding and domain knowledge. We show that LLMs can uncover latent intent behind these expressions in e-commerce queries through a framework that extracts structured interpretations or hints. Our approach decomposes queries into attribute-value hints generated concurrently with retrieval, enabling efficient integration into the ranking pipeline. Our method improves search performanc eby 10.9 points in MAP and ranking by 5.9 points in MRR over baselines. Since direct LLM-based reranking faces prohibitive latency, we develop an efficient approach transferring superlative interpretations to lightweight models. Our findings provide insights into how superlative semantics can be represented and transferred between models, advancing linguistic interpretation in retrieval systems while addressing practical deployment constraints.