Hierarchical Text Classification with LLM-Refined Taxonomies

• URL: http://arxiv.org/abs/2601.18375v1
• Date: Mon, 26 Jan 2026 11:28:32 GMT
• Title: Hierarchical Text Classification with LLM-Refined Taxonomies
• Authors: Jonas Golde, Nicolaas Jedema, Ravi Krishnan, Phong Le,
• Abstract summary: We present TaxMorph, a framework that uses large language models (LLMs) to transform entire operations such as renaming, merging, splitting, and reordering.<n>Experiments across three HTC benchmarks show that LLM-refined consistently outperform human-curated ones in various settings up to +2.9pp. in F1.
• Score: 1.5773713889677439
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Hierarchical text classification (HTC) depends on taxonomies that organize labels into structured hierarchies. However, many real-world taxonomies introduce ambiguities, such as identical leaf names under similar parent nodes, which prevent language models (LMs) from learning clear decision boundaries. In this paper, we present TaxMorph, a framework that uses large language models (LLMs) to transform entire taxonomies through operations such as renaming, merging, splitting, and reordering. Unlike prior work, our method revises the full hierarchy to better match the semantics encoded by LMs. Experiments across three HTC benchmarks show that LLM-refined taxonomies consistently outperform human-curated ones in various settings up to +2.9pp. in F1. To better understand these improvements, we compare how well LMs can assign leaf nodes to parent nodes and vice versa across human-curated and LLM-refined taxonomies. We find that human-curated taxonomies lead to more easily separable clusters in embedding space. However, the LLM-refined taxonomies align more closely with the model's actual confusion patterns during classification. In other words, even though they are harder to separate, they better reflect the model's inductive biases. These findings suggest that LLM-guided refinement creates taxonomies that are more compatible with how models learn, improving HTC performance.

Related papers

• Enriching Taxonomies Using Large Language Models [1.546945230112218]
  We present Taxoria, a novel taxonomy enrichment pipeline that leverages Large Language Models (LLMs) to enhance a given taxonomy.<n>Unlike approaches that extract internal LLM, Taxoria uses an existing taxonomy as a seed and prompts an LLM to propose candidate nodes for enrichment.<n>The final output includes an enriched taxonomy with provenance tracking and visualization of the final merged taxonomy for analysis.
  arXiv  Detail & Related papers  (2025-11-21T10:25:59Z)
• QUDsim: Quantifying Discourse Similarities in LLM-Generated Text [70.22275200293964]
  We introduce an abstraction based on linguistic theories in Questions Under Discussion (QUD) and question semantics to help quantify differences in discourse progression.<n>We then use this framework to build $textbfQUDsim$, a similarity metric that can detect discursive parallels between documents.<n>Using QUDsim, we find that LLMs often reuse discourse structures (more so than humans) across samples, even when content differs.
  arXiv  Detail & Related papers  (2025-04-12T23:46:09Z)
• Taxonomy Tree Generation from Citation Graph [15.188580557890942]
  HiGTL is a novel end-to-end framework guided by human-provided instructions or preferred topics.<n>We develop a novel taxonomy node verbalization strategy that iteratively generates central concepts for each cluster.<n>Experiments demonstrate that HiGTL effectively produces coherent, high-quality concept.
  arXiv  Detail & Related papers  (2024-10-02T13:02:03Z)
• Are Large Language Models a Good Replacement of Taxonomies? [25.963448807848746]
  Large language models (LLMs) demonstrate an impressive ability to internalize knowledge and answer natural language questions. We ask if the schema of knowledge graph (i.e., taxonomy) is made obsolete by LLMs.
  arXiv  Detail & Related papers  (2024-06-17T01:21:50Z)
• TaxoLLaMA: WordNet-based Model for Solving Multiple Lexical Semantic Tasks [54.01153045234468]
  In this paper, we explore the capabilities of LLMs in capturing lexical-semantic knowledge from WordNet. We present TaxoLLaMA, the everything-in-one model, lightweight due to 4-bit quantization and LoRA.
  arXiv  Detail & Related papers  (2024-03-14T09:21:25Z)
• Fantastic Semantics and Where to Find Them: Investigating Which Layers of Generative LLMs Reflect Lexical Semantics [50.982315553104975]
  We investigate the bottom-up evolution of lexical semantics for a popular large language model, namely Llama2. Our experiments show that the representations in lower layers encode lexical semantics, while the higher layers, with weaker semantic induction, are responsible for prediction. This is in contrast to models with discriminative objectives, such as mask language modeling, where the higher layers obtain better lexical semantics.
  arXiv  Detail & Related papers  (2024-03-03T13:14:47Z)
• TELEClass: Taxonomy Enrichment and LLM-Enhanced Hierarchical Text Classification with Minimal Supervision [41.05874642535256]
  Hierarchical text classification is a fundamental web text mining task with broad applications such as web content analysis and semantic indexing.<n>Most earlier works focus on fully or semi-supervised methods that require a large amount of human annotated data which is costly and time-consuming to acquire.<n>To alleviate human efforts, we work on hierarchical text classification with a minimal amount of supervision: using the sole class name of each node as the only supervision.
  arXiv  Detail & Related papers  (2024-02-29T22:26:07Z)
• Using Zero-shot Prompting in the Automatic Creation and Expansion of Topic Taxonomies for Tagging Retail Banking Transactions [0.0]
  This work presents an unsupervised method for constructing and expanding topic using instruction-based fine-tuned LLMs (Large Language Models) To expand an existing taxonomy with new terms, we use zero-shot prompting to find out where to add new nodes. We use the resulting tags to assign tags that characterize merchants from a retail bank dataset.
  arXiv  Detail & Related papers  (2024-01-08T00:27:16Z)
• Octet: Online Catalog Taxonomy Enrichment with Self-Supervision [67.26804972901952]
  We present a self-supervised end-to-end framework, Octet for Online Catalog EnrichmenT. We propose to train a sequence labeling model for term extraction and employ graph neural networks (GNNs) to capture the taxonomy structure. Octet enriches an online catalog in production to 2 times larger in the open-world evaluation.
  arXiv  Detail & Related papers  (2020-06-18T04:53:07Z)
• STEAM: Self-Supervised Taxonomy Expansion with Mini-Paths [53.45704816829921]
  We propose a self-supervised taxonomy expansion model named STEAM. STEAM generates natural self-supervision signals, and formulates a node attachment prediction task. Experiments show STEAM outperforms state-of-the-art methods for taxonomy expansion by 11.6% in accuracy and 7.0% in mean reciprocal rank.
  arXiv  Detail & Related papers  (2020-06-18T00:32:53Z)
• TaxoExpan: Self-supervised Taxonomy Expansion with Position-Enhanced Graph Neural Network [62.12557274257303]
  Taxonomies consist of machine-interpretable semantics and provide valuable knowledge for many web applications. We propose a novel self-supervised framework, named TaxoExpan, which automatically generates a set of query concept, anchor concept> pairs from the existing taxonomy as training data. We develop two innovative techniques in TaxoExpan: (1) a position-enhanced graph neural network that encodes the local structure of an anchor concept in the existing taxonomy, and (2) a noise-robust training objective that enables the learned model to be insensitive to the label noise in the self-supervision data.
  arXiv  Detail & Related papers  (2020-01-26T21:30:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.