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

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhu, Yilun, Vedula, Nikhita, Malmasi, Shervin
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911271671037952
author Zhu, Yilun
Vedula, Nikhita
Malmasi, Shervin
author_facet Zhu, Yilun
Vedula, Nikhita
Malmasi, Shervin
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_13994
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hint-Augmented Re-ranking: Efficient Product Search using LLM-Based Query Decomposition
Zhu, Yilun
Vedula, Nikhita
Malmasi, Shervin
Computation and Language
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.
title Hint-Augmented Re-ranking: Efficient Product Search using LLM-Based Query Decomposition
topic Computation and Language
url https://arxiv.org/abs/2511.13994