Improving Product Search Relevance with EAR-MP: A Solution for the CIKM 2025 AnalytiCup

TL;DR

This paper presents EAR-MP, a multilingual e-commerce search solution for the CIKM 2025 AnalytiCup, focusing on improving query relevance through data normalization, advanced training techniques, and task-specific enhancements, achieving high F1 scores.

Contribution

The paper introduces a comprehensive approach combining data normalization, model improvements, and task-specific strategies for multilingual relevance in e-commerce search.

Findings

Achieved F1 score of 0.8796 on Query-Category relevance

Attained F1 score of 0.8744 on Query-Item relevance

Demonstrated the effectiveness of systematic data preprocessing and tailored training.

Abstract

Multilingual e-commerce search is challenging due to linguistic diversity and the noise inherent in user-generated queries. This paper documents the solution employed by our team (EAR-MP) for the CIKM 2025 AnalytiCup, which addresses two core tasks: Query-Category (QC) relevance and Query-Item (QI) relevance. Our approach first normalizes the multilingual dataset by translating all text into English, then mitigates noise through extensive data cleaning and normalization. For model training, we build on DeBERTa-v3-large and improve performance with label smoothing, self-distillation, and dropout. In addition, we introduce task-specific upgrades, including hierarchical token injection for QC and a hybrid scoring mechanism for QI. Under constrained compute, our method achieves competitive results, attaining an F1 score of 0.8796 on QC and 0.8744 on QI. These findings underscore the importance of systematic data preprocessing and tailored training strategies for building robust, resource-efficient multilingual relevance systems.