AI-PAVE-Br: Leveraging Large Language Models for Enhanced Product Attribute Value Extraction through a Golden Set Approach

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Murilo Gazzola, Hugo Gobato Souto, Samuel Silva, Júlia Schubert Peixoto, Felipe Siqueira, André Luis Pedroso de Morais, Caio Gomes
• Venue: arXiv: cs.CL
• Published: 2026-06-23
• Collected: 2026-06-24
• URL: http://arxiv.org/abs/2606.24655v1
• PDF: https://arxiv.org/pdf/2606.24655v1
• DOI: 10.5753/stil.2025.37821

Plain-Language Brief

The paper presents AI-PAVEBr, a system that uses large language models to extract product attribute values from Brazilian e-commerce catalogs. It also introduces the Golden Set, a new dataset in Portuguese for product attribute value extraction (PAVE), which will aid in reproducible research and provide a benchmark for future studies.

What the Document Says

• The paper discusses a specialized system, AI-PAVEBr, designed for high-accuracy Product Attribute Value Extraction (PAVE) specifically for Brazilian e-commerce catalogs using Large Language Models (LLMs).
• A new dataset called the Golden Set is introduced. This dataset is meticulously curated and manually annotated in Portuguese for PAVE.
• The Golden Set consists of Entities, Categories, and Subcategories.
• The authors claim that AI-PAVE-Br outperforms conventional Named Entity Recognition (NER) baselines when leveraging targeted prompt engineering.

Key Concepts Mentioned

• Product Attribute Value Extraction
• Large Language Models
• Brazilian e-commerce catalogs
• Golden Set
• Named Entity Recognition

Claims Made by the Document

• AI-PAVEBr outperforms conventional Named Entity Recognition (NER) baselines.
• The Golden Set is a new, high-quality reference set for PAVE in Portuguese.

Evidence Used

• The experiments conclusively show that AI-PAVE-Br outperforms conventional NER baselines.

Methods / Approach

• Large Language Models are used for Product Attribute Value Extraction.
• The Golden Set is created for PAVE in Portuguese.

Limitations Stated by the Authors

Not stated in source

Limitations Noted by the Agent

Agent interpretation — not from the source.

The paper does not discuss any potential limitations of the AI-PAVEBr system or the Golden Set.

Open Questions

• What are the specific improvements in accuracy that AI-PAVEBr offers over conventional NER baselines?
• How diverse and comprehensive is the Golden Set in terms of product categories and descriptions?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The paper provides evidence that AI-PAVEBr outperforms conventional NER baselines, but it does not detail the specific improvements.

Raw Abstract / Excerpt

The explosive growth and complexity of product data within the dynamic Brazilian e-commerce landscape demand robust and specialized methods for structured information extraction. Traditional approaches to Product Attribute Value Extraction (PAVE) often struggle with the linguistic nuances and sheer diversity of product descriptions in Portuguese. To address this critical gap, this paper introduces two major contributions. First, we present AI-PAVEBr, a specialized system engineered with Large Language Models (LLMs) to perform high-accuracy PAVE specifically for Brazilian e-commerce catalogs. Second, to facilitate reproducible research and provide a definitive benchmark, we introduce and share the Golden Set, a new, meticulously curated, and manually annotated dataset for PAVE in Portuguese. We detail the creation process and structure (Entity, Category, Subcategories) of this high-quality reference set. Our experiments conclusively show that AI-PAVE-Br, leveraging targeted prompt engineering, dramatically outperforms conventional Named Entity Recognition (NER) baselines. This work not only delivers a superior, scalable solution for a major non-English market but also enriches the NLP community with a valuable, publicly available resource for future PAVE research.