Hint-Augmented Re-ranking: Efficient Product Search using LLM-Based Query Decomposition
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 18 of its 32 references that are in the map. First posted on arXiv 17 Nov 2025.
Builds on
18 of 32 references.
References · 18
- 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
- 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,605
- 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 853
- 2023Towards General Text Embeddings with Multi-stage Contrastive Learningcited by 723
- 2023Large Language Models are Effective Text Rankers with Pairwise Ranking Promptingcited by 307
- 2023A Setwise Approach for Effective and Highly Efficient Zero-shot Ranking with Large Language Modelscited by 55
- 2024Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankerscited by 43
- 2025The Amazon Nova Family of Models: Technical Report and Model Cardcited by 41
- 2024JudgeRank: Leveraging Large Language Models for Reasoning-Intensive Rerankingcited by 34
- 2024CALRec: Contrastive Alignment of Generative LLMs for Sequential Recommendationcited by 24
- 2024GenQREnsemble: Zero-Shot LLM Ensemble Prompting for Generative Query Reformulationcited by 23
- 2025Wizard of Shopping: Target-Oriented E-commerce Dialogue Generation with Decision Tree Branchingcited by 10
6 more
- 2024Question Suggestion for Conversational Shopping Assistants Using Product Metadatacited by 8
- 2024Generative Explore-Exploit: Training-free Optimization of Generative Recommender Systems using LLM Optimizerscited by 7
- 2024Superlatives in Context: Modeling the Implicit Semantics of Superlativescited by 5
- 2025Generative Product Recommendations for Implicit Superlative Queriescited by 5
- 2024Bridging the Gap Between Information Seeking and Product Search Systems: Q&A Recommendation for E-commercecited by 4
- 2024Identifying High Consideration E-Commerce Search Queriescited by 4
Built on by
Citing papers per year: 2026: 1.
1 paper.