Generative Product Recommendations for Implicit Superlative Queries

2025-04-26

AI Summary

Traditional e-commerce recommendation systems struggle with implicit superlative queries—e.g., “best shoes for trail running”—that lack explicitly stated comparative attributes, resulting in insufficient attribute signals for modeling. Method: We propose SUPERB, a novel four-dimensional annotation framework that systematically constructs implicit superlative product candidate sets for the first time. To bridge semantic ambiguity, we design an LLM-driven approach for implicit attribute generation and reasoning, transforming unstructured natural language into structured, interpretable attribute representations. Contribution/Results: Evaluated on a high-quality, manually annotated dataset via joint retrieval-and-ranking metrics, our method achieves significant gains in recommendation accuracy. This work introduces the first production-ready, generative attribute-enhancement paradigm for industrial e-commerce recommendation, advancing implicit intent modeling from discriminative to generative paradigms.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Natural Language Processing: Generation
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

In Recommender Systems, users often seek the best products through indirect, vague, or under-specified queries, such as"best shoes for trail running". Such queries, also referred to as implicit superlative queries, pose a significant challenge for standard retrieval and ranking systems as they lack an explicit mention of attributes and require identifying and reasoning over complex factors. We investigate how Large Language Models (LLMs) can generate implicit attributes for ranking as well as reason over them to improve product recommendations for such queries. As a first step, we propose a novel four-point schema for annotating the best product candidates for superlative queries called SUPERB, paired with LLM-based product annotations. We then empirically evaluate several existing retrieval and ranking approaches on our new dataset, providing insights and discussing their integration into real-world e-commerce production systems.

Problem

• Addressing vague user queries in recommender systems
• Using LLMs to infer attributes for product ranking
• Improving recommendations for implicit superlative queries

Innovation

• LLMs generate implicit attributes for ranking
• Propose SUPERB schema for product annotation
• Evaluate retrieval methods on new dataset