TACLR: A Scalable and Efficient Retrieval-based Method for Industrial Product Attribute Value Identification

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Su, Yindu, Zou, Huike, Sun, Lin, Zhang, Ting, Yang, Haiyang, Chen, Liyu, Lo, David, Zhang, Qingheng, Han, Shuguang, Chen, Jufeng
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910981532155904
author Su, Yindu
Zou, Huike
Sun, Lin
Zhang, Ting
Yang, Haiyang
Chen, Liyu
Lo, David
Zhang, Qingheng
Han, Shuguang
Chen, Jufeng
author_facet Su, Yindu
Zou, Huike
Sun, Lin
Zhang, Ting
Yang, Haiyang
Chen, Liyu
Lo, David
Zhang, Qingheng
Han, Shuguang
Chen, Jufeng
contents Product Attribute Value Identification (PAVI) involves identifying attribute values from product profiles, a key task for improving product search, recommendation, and business analytics on e-commerce platforms. However, existing PAVI methods face critical challenges, such as inferring implicit values, handling out-of-distribution (OOD) values, and producing normalized outputs. To address these limitations, we introduce Taxonomy-Aware Contrastive Learning Retrieval (TACLR), the first retrieval-based method for PAVI. TACLR formulates PAVI as an information retrieval task by encoding product profiles and candidate values into embeddings and retrieving values based on their similarity. It leverages contrastive training with taxonomy-aware hard negative sampling and employs adaptive inference with dynamic thresholds. TACLR offers three key advantages: (1) it effectively handles implicit and OOD values while producing normalized outputs; (2) it scales to thousands of categories, tens of thousands of attributes, and millions of values; and (3) it supports efficient inference for high-load industrial deployment. Extensive experiments on proprietary and public datasets validate the effectiveness and efficiency of TACLR. Further, it has been successfully deployed on the real-world e-commerce platform Xianyu, processing millions of product listings daily with frequently updated, large-scale attribute taxonomies. We release the code to facilitate reproducibility and future research at https://github.com/SuYindu/TACLR.
format Preprint
id arxiv_https___arxiv_org_abs_2501_03835
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TACLR: A Scalable and Efficient Retrieval-based Method for Industrial Product Attribute Value Identification
Su, Yindu
Zou, Huike
Sun, Lin
Zhang, Ting
Yang, Haiyang
Chen, Liyu
Lo, David
Zhang, Qingheng
Han, Shuguang
Chen, Jufeng
Computation and Language
Artificial Intelligence
Information Retrieval
Product Attribute Value Identification (PAVI) involves identifying attribute values from product profiles, a key task for improving product search, recommendation, and business analytics on e-commerce platforms. However, existing PAVI methods face critical challenges, such as inferring implicit values, handling out-of-distribution (OOD) values, and producing normalized outputs. To address these limitations, we introduce Taxonomy-Aware Contrastive Learning Retrieval (TACLR), the first retrieval-based method for PAVI. TACLR formulates PAVI as an information retrieval task by encoding product profiles and candidate values into embeddings and retrieving values based on their similarity. It leverages contrastive training with taxonomy-aware hard negative sampling and employs adaptive inference with dynamic thresholds. TACLR offers three key advantages: (1) it effectively handles implicit and OOD values while producing normalized outputs; (2) it scales to thousands of categories, tens of thousands of attributes, and millions of values; and (3) it supports efficient inference for high-load industrial deployment. Extensive experiments on proprietary and public datasets validate the effectiveness and efficiency of TACLR. Further, it has been successfully deployed on the real-world e-commerce platform Xianyu, processing millions of product listings daily with frequently updated, large-scale attribute taxonomies. We release the code to facilitate reproducibility and future research at https://github.com/SuYindu/TACLR.
title TACLR: A Scalable and Efficient Retrieval-based Method for Industrial Product Attribute Value Identification
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2501.03835