Taxonomy Exploration with Large Language Model Reasoning and Tool Calling
Main Article Content
Abstract
The paper addresses the task of taxonomy expansion – hierarchical structures for organizing concepts. We propose an architecture based on the ReAct (Reasoning + Acting) approach that enables taxonomy expansion in a zero-shot setting without fine-tuning large language models. The system is implemented in two scenarios: autonomous navigation from root nodes and verification of hypotheses generated by other models. Experiments on the diachronic RuWordNet dataset show that direct navigation from the root faces limitations due to graph complexity (MAP@3 = 24.6%). However, using the system as a verifier improves the performance of baseline models: MAP@3 gains of 9.5 pp for FastText and 1.1 pp for TaxoYandexGPT-5-Lite. The key advantages of the approach are its universality, the absence of fine-tuning requirements, and interpretability through explicit reasoning chains.
Article Details
References
2. Xu H. et al. Compress and mix: Advancing efficient taxonomy completion with large language models // Proceedings of the ACM on Web Conference 2025. 2025. P. 4239–4249. https://doi.org/10.1145/3696410.3714690
3. Niu Y. et al. Contrastive representation learning for self-supervised taxonomy completion // IJCAI. 2024. Vol. 8. P. 6442–6450. https://doi.org/10.24963/ijcai.2024/712
4. Arous I., Dolamic L., Cudré-Mauroux P. Taxocomplete: Self-supervised taxonomy completion leveraging position-enhanced semantic matching //Proceedings of the ACM Web Conference 2023. 2023. P. 2509–2518. https://doi.org/10.1145/3543507.3583342
5. Xu H. et al. Tacoprompt: A collaborative multi-task prompt learning method for self-supervised taxonomy completion //Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023. P. 15804–15817. https://doi.org/10.18653/v1/2023.emnlp-main.979
6. D'Amico S. et al. Taxonomy Expansion through Collaborative LLM Mapping //Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing. 2025. P. 1961–1968. https://doi.org/10.1145/3672608.3707906
7. Moskvoretskii V. et al. Large Language Models for Creation, Enrichment and Evaluation of Taxonomic Graphs // Semantic Web: Interoperability, Usability, Applicability. 2026. Vol. 17, No. 1. https://doi.org/10.1177/22104968251404186
8. Yao S. et al. ReAct: Synergizing Reasoning and Acting in Language Models // International Conference on Learning Representations (ICLR). 2023. URL: https://openreview.net/forum?id=WE_vluYUL-X (Accessed: 31.03.2026).
9. Nikishina I. et al. Russe'2020: Findings of the first taxonomy enrichment task for the Russian language // Computational Linguistics and Intellectual Technologies. 2020. P. 579–595. URL: https://api.semanticscholar.org/CorpusID:218862855 (Accessed: 31.03.2026).
10. Nikishina I. et al. Studying taxonomy enrichment on diachronic WordNet versions // Proceedings of the 28th International Conference on Computational Linguistics. 2020. P. 3095–3106. https://doi.org/10.18653/v1/2020.coling-main.276
11. Miller G.A. et al. Introduction to WordNet: An on-line lexical database // International journal of lexicography. 1990. Vol. 3, No. 4. P. 235–244.
12. Loukachevitch N.V. et al. Creating Russian wordnet by conversion // Computational Linguistics and Intellectual Technologies. 2016. P. 405–415. URL: https://api.semanticscholar.org/CorpusID:63871134 (Accessed: 31.03.2026).
13. Nikishina I. et al. Taxonomy enrichment with text and graph vector representations // Semantic Web. 2022. Vol. 13, No. 3. P. 441–475. https://doi.org/10.1007/978-3-031-16500-9_2
14. Tikhomirov M., Loukachevitch N. Meta-embeddings in taxonomy enrichment task // Computational Linguistics and Intellectual Technologies: papers from the Annual conference “Dialogue. 2021. URL: https://dialogue-conf.org/media/5549/tikhomirovmmplusloukachevitchnv091.pdf (Accessed: 31.03.2026).
This work is licensed under a Creative Commons Attribution 4.0 International License.
Presenting an article for publication in the Russian Digital Libraries Journal (RDLJ), the authors automatically give consent to grant a limited license to use the materials of the Kazan (Volga) Federal University (KFU) (of course, only if the article is accepted for publication). This means that KFU has the right to publish an article in the next issue of the journal (on the website or in printed form), as well as to reprint this article in the archives of RDLJ CDs or to include in a particular information system or database, produced by KFU.
All copyrighted materials are placed in RDLJ with the consent of the authors. In the event that any of the authors have objected to its publication of materials on this site, the material can be removed, subject to notification to the Editor in writing.
Documents published in RDLJ are protected by copyright and all rights are reserved by the authors. Authors independently monitor compliance with their rights to reproduce or translate their papers published in the journal. If the material is published in RDLJ, reprinted with permission by another publisher or translated into another language, a reference to the original publication.
By submitting an article for publication in RDLJ, authors should take into account that the publication on the Internet, on the one hand, provide unique opportunities for access to their content, but on the other hand, are a new form of information exchange in the global information society where authors and publishers is not always provided with protection against unauthorized copying or other use of materials protected by copyright.
RDLJ is copyrighted. When using materials from the log must indicate the URL: index.phtml page = elbib / rus / journal?. Any change, addition or editing of the author's text are not allowed. Copying individual fragments of articles from the journal is allowed for distribute, remix, adapt, and build upon article, even commercially, as long as they credit that article for the original creation.
Request for the right to reproduce or use any of the materials published in RDLJ should be addressed to the Editor-in-Chief A.M. Elizarov at the following address: amelizarov@gmail.com.
The publishers of RDLJ is not responsible for the view, set out in the published opinion articles.
We suggest the authors of articles downloaded from this page, sign it and send it to the journal publisher's address by e-mail scan copyright agreements on the transfer of non-exclusive rights to use the work.