Hierarchical Text Classification Using Contrastive Learning Informed Path Guided Hierarchy

2025-06-04

European Conference on Artificial Intelligence

AI Summary

Hierarchical text classification suffers from a disconnect between semantic modeling and label structure exploitation. Method: This paper proposes a dual-stream contrastive learning framework that jointly models label hierarchies and textual semantics. It is the first to integrate path-guided hierarchical representation learning with contrastive learning, establishing a unified dual-branch probabilistic output mechanism that synergistically combines path-based hierarchy modeling and semantics-aware text encoding within a single architecture. Joint optimization of the text encoder and hierarchy encoder enforces consistency alignment between hierarchy-aware and path-guided representations. Results: The method achieves state-of-the-art performance, improving Macro F1 by 0.99–2.37% over prior approaches on two benchmark datasets. Its core contribution is the first unified dual-stream paradigm integrating path guidance and contrastive learning—effectively bridging the semantic–structural modeling gap inherent in conventional hierarchical classification.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Hierarchical Text Classification (HTC) has recently gained traction given the ability to handle complex label hierarchy. This has found applications in domains like E- commerce, customer care and medicine industry among other real-world applications. Existing HTC models either encode label hierarchy separately and mix it with text encoding or guide the label hierarchy structure in the text encoder. Both approaches capture different characteristics of label hierarchy and are complementary to each other. In this paper, we propose a Hierarchical Text Classification using Contrastive Learning Informed Path guided hierarchy (HTC-CLIP), which learns hierarchy-aware text representation and text informed path guided hierarchy representation using contrastive learning. During the training of HTC-CLIP, we learn two different sets of class probabilities distributions and during inference, we use the pooled output of both probabilities for each class to get the best of both representations. Our results show that the two previous approaches can be effectively combined into one architecture to achieve improved performance. Tests on two public benchmark datasets showed an improvement of 0.99 - 2.37% in Macro F1 score using HTC-CLIP over the existing state-of-the-art models.

Problem

• Combining label hierarchy and text encoding for hierarchical text classification
• Improving hierarchical text classification with contrastive learning
• Enhancing performance by pooling complementary probability distributions

Innovation

• Combines hierarchy-aware text and label representations
• Uses contrastive learning for dual representation training
• Pools dual probabilities for improved classification performance