# HILL: Hierarchy-aware Information Lossless Contrastive Learning for Hierarchical Text Classification

arXiv: https://arxiv.org/abs/2403.17307 · HTML: https://mlsanity.com/paper/2403.17307/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Mar 2024.

## Built on by (most cited first, 13 of 13)

- [Revisiting Hierarchical Text Classification: Inference and Metrics](https://mlsanity.com/paper/2410.01305/) (2024)
- [NC-NCD: Novel Class Discovery for Node Classification](https://mlsanity.com/paper/2407.17816/) (2024)
- [Rumor Detection on Social Media with Temporal Propagation Structure Optimization](https://mlsanity.com/paper/2412.08316/) (2024)
- [Your Extreme Multi-label Classifier is Secretly a Hierarchical Text Classifier for Free](https://mlsanity.com/paper/2411.13687/) (2024)
- [Hierarchical Text Classification with LLM-Refined Taxonomies](https://mlsanity.com/paper/2601.18375/) (2026)
- [Correcting Noisy Multilabel Predictions: Modeling Label Noise through Latent Space Shifts](https://mlsanity.com/paper/2502.14281/) (2025)
- [Uncovering Capabilities of Model Pruning in Graph Contrastive Learning](https://mlsanity.com/paper/2410.20356/) (2024)
- [Adaptive Data-Resilient Multi-Modal Hierarchical Multi-Label Book Genre Identification](https://mlsanity.com/paper/2505.03839/) (2025)
- [Toward Robust Signed Graph Learning through Joint Input-Target Denoising](https://mlsanity.com/paper/2510.22513/) (2025)
- [Hierarchical Book Organization for Learning-Resource Discovery using Dual-Path Graph Convolutions](https://mlsanity.com/paper/2512.21076/) (2025)
- [Taxon: Hierarchical Tax Code Prediction with Semantically Aligned LLM Expert Guidance](https://mlsanity.com/paper/2601.08418/) (2026)
- [Mastering the Minority: An Uncertainty-guided Multi-Expert Framework for Challenging-tailed Sequence Learning](https://mlsanity.com/paper/2603.15708/) (2026)
- [Semantic Reranking at Inference Time for Hard Examples in Rhetorical Role Labeling](https://mlsanity.com/paper/2605.18007/) (2026)

## Abstract (arXiv)

Existing self-supervised methods in natural language processing (NLP), especially hierarchical text classification (HTC), mainly focus on self-supervised contrastive learning, extremely relying on human-designed augmentation rules to generate contrastive samples, which can potentially corrupt or distort the original information. In this paper, we tend to investigate the feasibility of a contrastive learning scheme in which the semantic and syntactic information inherent in the input sample is adequately reserved in the contrastive samples and fused during the learning process. Specifically, we propose an information lossless contrastive learning strategy for HTC, namely Hierarchy-aware Information Lossless contrastive Learning (HILL), which consists of a text encoder representing the input document, and a structure encoder directly generating the positive sample. The structure encoder takes the document embedding as input, extracts the essential syntactic information inherent in the label hierarchy with the principle of structural entropy minimization, and injects the syntactic information into the text representation via hierarchical representation learning. Experiments on three common datasets are conducted to verify the superiority of HILL.