BEATS: Bootstrapping E-commerce Attribute Taxonomies for Search through Iterative Human-AI Collaboration

• URL: http://arxiv.org/abs/2606.04909v1
• Date: Wed, 03 Jun 2026 14:06:27 GMT
• Title: BEATS: Bootstrapping E-commerce Attribute Taxonomies for Search through Iterative Human-AI Collaboration
• Abstract summary: We present BEATS, a human-in-the-loop LLM framework for bootstrapping product attribute entirely from scratch.<n>Our system has been deployed at Rakuten Taiwan, enriching 9 major categories spanning 2,694 sub-categories with 67,277 generated attributes.<n>Over 5.4 million products have been tagged with the generated attributes, with plans to enrich the entire product catalog.
• Score: 16.653517324875384
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: E-commerce platforms in emerging markets often operate with underdeveloped product catalogs that contain only category taxonomies but lack structured attribute schemas. This absence of fine-grained product attributes limits search capabilities -- preventing faceted filtering, degrading query understanding, and weakening semantic representations used by search systems. We present BEATS, a human-in-the-loop LLM framework for bootstrapping product attribute taxonomies entirely from scratch. Our approach extends a multi-stage LLM generation pipeline with two critical production stages: (1) proactive quality checking by model developers to filter erroneous outputs, and (2) human annotation by domain-expert local staff to validate generated attributes. The framework operates iteratively -- prompts at each generation stage are refined based on quality check observations and annotator feedback across successive rounds, progressively improving attribute quality. Once the attribute taxonomy is established, we employ LLMs to perform structured attribute tagging on individual product items, enriching their contextual representations. The enriched catalog directly benefits multiple components of the search system: enabling granular attribute-based filtering, providing structured features for ranking models, and improving semantic representations for dense retrieval. We validate the generated taxonomy by training dense retrieval models on attribute-enriched product data, demonstrating consistent improvements over baselines using original catalog information. Our system has been deployed at Rakuten Taiwan, enriching 9 major categories spanning 2,694 sub-categories with 67,277 generated attributes, and over 5.4 million products have been tagged with the generated attributes, with plans to enrich the entire product catalog.

Related papers

• Scaling E-Commerce Attribute Extraction with Parallel Decoding [25.167068017765768]
  e-commerce catalogs are messy and unstructured, making it difficult to identify which attributes matter most.<n>We introduce a two-stage LLM pipeline that first discovers a compact, ranked schema of purchase-discriminative attributes for each product category.<n>This pipeline achieves 85% extraction accuracy, on par with the foundational LLM it was distilled from, while reducing inference costs by 92%.
  arXiv  Detail & Related papers  (2026-09-09T05:01:28Z)
• TRACE: Agentic Catalog Enrichment with Multi-source Evidence Grounding [0.8677773374912304]
  This paper introduces TRACE, a novel framework for automated catalog attribute enrichment using agentic Large Language Models (LLMs)<n>On an offline human evaluation dataset, TRACE's proposed attribute values were 98.2% accurate at 74.7% attribute coverage.<n>An online experiment subsequently showed that surfacing the enriched attributes on the product detail page increased checkout conversion by 0.48%.
  arXiv  Detail & Related papers  (2026-08-21T08:08:27Z)
• From Unstructured to Structured: LLM-Guided Attribute Graphs for Entity Search and Ranking [20.966359103135762]
  We present a two-stage approach combining Large Language Model (LLM)-driven attribute graph construction with graph-aware LLM ranking.<n>In the offline stage, we extract structured product attributes from unstructured text, and construct a reusable attribute graph with category-aware schemas.<n>In the online stage, we rank retrieved candidates by reasoning over this structured representation rather than raw text, reducing per-product token usage by 57% while improving ranking precision.
  arXiv  Detail & Related papers  (2026-04-30T04:22:20Z)
• Multi-Value-Product Retrieval-Augmented Generation for Industrial Product Attribute Value Identification [12.167857724257239]
  We introduce Multi-Value-Product Retrieval-Augmented Generation (MVP-RAG)<n>MVP-RAG defines PAVI as a retrieval-generation task, where the product title description serves as the query.<n>It first retrieves similar products of the same category and candidate attribute values, and then generates the standardized attribute values.
  arXiv  Detail & Related papers  (2025-09-28T13:29:20Z)
• Self-Refinement Strategies for LLM-based Product Attribute Value Extraction [51.45146101802871]
  This paper investigates applying two self-refinement techniques to the product attribute value extraction task.<n>The experiments show that both self-refinement techniques fail to significantly improve the extraction performance while substantially increasing processing costs.<n>For scenarios with development data, fine-tuning yields the highest performance, while the ramp-up costs of fine-tuning are balanced out as the amount of product descriptions increases.
  arXiv  Detail & Related papers  (2025-01-02T12:55:27Z)
• Large Language Models for Relevance Judgment in Product Search [48.56992980315751]
  High relevance of retrieved and re-ranked items to the search query is the cornerstone of successful product search. We present an array of techniques for leveraging Large Language Models (LLMs) for automating the relevance judgment of query-item pairs (QIPs) at scale. Our findings have immediate implications for the growing field of relevance judgment automation in product search.
  arXiv  Detail & Related papers  (2024-06-01T00:52:41Z)
• Using LLMs for the Extraction and Normalization of Product Attribute Values [47.098255866050835]
  This paper explores the potential of using large language models (LLMs) to extract and normalize attribute values from product titles and descriptions. We introduce the Web Data Commons - Product Attribute Value Extraction (WDC-PAVE) benchmark dataset for our experiments.
  arXiv  Detail & Related papers  (2024-03-04T15:39:59Z)
• Enhanced E-Commerce Attribute Extraction: Innovating with Decorative Relation Correction and LLAMA 2.0-Based Annotation [4.81846973621209]
  We propose a pioneering framework that integrates BERT for classification, a Conditional Random Fields (CRFs) layer for attribute value extraction, and Large Language Models (LLMs) for data annotation. Our approach capitalizes on the robust representation learning of BERT, synergized with the sequence decoding prowess of CRFs, to adeptly identify and extract attribute values. Our methodology is rigorously validated on various datasets, including Walmart, BestBuy's e-commerce NER dataset, and the CoNLL dataset.
  arXiv  Detail & Related papers  (2023-12-09T08:26:30Z)
• ExtractGPT: Exploring the Potential of Large Language Models for Product Attribute Value Extraction [51.87391234815163]
  E-commerce platforms require structured product data in the form of attribute-value pairs.<n>BERT-based extraction methods require large amounts of task-specific training data.<n>This paper explores using large language models (LLMs) as a more training-data efficient and robust alternative.
  arXiv  Detail & Related papers  (2023-10-19T07:39:00Z)
• Product Information Extraction using ChatGPT [69.12244027050454]
  This paper explores the potential of ChatGPT for extracting attribute/value pairs from product descriptions. Our results show that ChatGPT achieves a performance similar to a pre-trained language model but requires much smaller amounts of training data and computation for fine-tuning.
  arXiv  Detail & Related papers  (2023-06-23T09:30:01Z)
• A Unified Generative Approach to Product Attribute-Value Identification [6.752749933406399]
  We explore a generative approach to the product attribute-value identification (PAVI) task. We finetune a pre-trained generative model, T5, to decode a set of attribute-value pairs as a target sequence from the given product text. Experimental results confirm that our generation-based approach outperforms the existing extraction and classification-based methods.
  arXiv  Detail & Related papers  (2023-06-09T00:33:30Z)
• OA-Mine: Open-World Attribute Mining for E-Commerce Products with Weak Supervision [93.26737878221073]
  We study the attribute mining problem in an open-world setting to extract novel attributes and their values. We propose a principled framework that first generates attribute value candidates and then groups them into clusters of attributes. Our model significantly outperforms strong baselines and can generalize to unseen attributes and product types.
  arXiv  Detail & Related papers  (2022-04-29T04:16:04Z)
• Automatic Validation of Textual Attribute Values in E-commerce Catalog by Learning with Limited Labeled Data [61.789797281676606]
  We propose a novel meta-learning latent variable approach, called MetaBridge. It can learn transferable knowledge from a subset of categories with limited labeled data. It can capture the uncertainty of never-seen categories with unlabeled data.
  arXiv  Detail & Related papers  (2020-06-15T21:31:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.