Context-Aware Hierarchical Taxonomy Generation for Scientific Papers via LLM-Guided Multi-Aspect Clustering

2025-09-23

AI Summary

To address the lack of coherence and fine-grained structure in classification schemes amid the explosive growth of scientific literature, this paper proposes an LLM-guided, context-aware hierarchical classification framework. Methodologically, it introduces a multi-dimensional paper representation—encompassing methods, datasets, evaluation protocols, and other key aspects—and leverages large language models for dimension-specific semantic parsing and abstractive summarization. These representations are then integrated via dynamic hierarchical clustering to enable cross-dimensional joint modeling. Key contributions include: (1) the first benchmark comprising 156 manually curated classification schemas covering 11.6K papers; and (2) state-of-the-art performance across three core metrics—coherence, granularity, and interpretability—outperforming all existing approaches. The framework thus advances automated, principled taxonomy construction for large-scale scientific corpora.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Knowledge Representation and Reasoning: Automated Reasoning and Theorem Proving
• Natural Language Processing: Summarization

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

The rapid growth of scientific literature demands efficient methods to organize and synthesize research findings. Existing taxonomy construction methods, leveraging unsupervised clustering or direct prompting of large language models (LLMs), often lack coherence and granularity. We propose a novel context-aware hierarchical taxonomy generation framework that integrates LLM-guided multi-aspect encoding with dynamic clustering. Our method leverages LLMs to identify key aspects of each paper (e.g., methodology, dataset, evaluation) and generates aspect-specific paper summaries, which are then encoded and clustered along each aspect to form a coherent hierarchy. In addition, we introduce a new evaluation benchmark of 156 expert-crafted taxonomies encompassing 11.6k papers, providing the first naturally annotated dataset for this task. Experimental results demonstrate that our method significantly outperforms prior approaches, achieving state-of-the-art performance in taxonomy coherence, granularity, and interpretability.

Problem

• Organizing rapidly growing scientific literature efficiently
• Improving coherence and granularity in taxonomy construction
• Generating context-aware hierarchical taxonomies for papers

Innovation

• LLM-guided multi-aspect encoding
• Dynamic clustering for hierarchy formation
• Context-aware aspect-specific paper summarization