Taxon: Hierarchical Tax Code Prediction with Semantically Aligned LLM Expert Guidance

TL;DR

Taxon is a novel framework that combines semantic alignment and expert guidance to improve hierarchical tax code prediction accuracy for large-scale e-commerce, with successful deployment in Alibaba.

Contribution

Introduces a semantically aligned, expert-guided hierarchical tax code prediction model integrating multi-modal features and large language model distillation.

Findings

Achieves state-of-the-art performance on proprietary and public benchmarks.

Significantly improves structural consistency in tax code predictions.

Successfully deployed in Alibaba's production system, handling over 500,000 queries daily.

Abstract

Tax code prediction is a crucial yet underexplored task in automating invoicing and compliance management for large-scale e-commerce platforms. Each product must be accurately mapped to a node within a multi-level taxonomic hierarchy defined by national standards, where errors lead to financial inconsistencies and regulatory risks. This paper presents Taxon, a semantically aligned and expert-guided framework for hierarchical tax code prediction. Taxon integrates (i) a feature-gating mixture-of-experts architecture that adaptively routes multi-modal features across taxonomy levels, and (ii) a semantic consistency model distilled from large language models acting as domain experts to verify alignment between product titles and official tax definitions. To address noisy supervision in real business records, we design a multi-source training pipeline that combines curated tax databases, invoice validation logs, and merchant registration data to provide both structural and semantic supervision. Extensive experiments on the proprietary TaxCode dataset and public benchmarks demonstrate that Taxon achieves state-of-the-art performance, outperforming strong baselines. Further, an additional full hierarchical paths reconstruction procedure significantly improves structural consistency, yielding the highest overall F1 scores. Taxon has been deployed in production within Alibaba's tax service system, handling an average of over 500,000 tax code queries per day and reaching peak volumes above five million requests during business event with improved accuracy, interpretability, and robustness.