LLM Confidence in Constructing Semantic Classifications of Arguments
Main Article Content
Abstract
This article addresses the problem of quantifying the confidence of large language models (LLMs) in the automatic semantic classification of arguments with emotive predicates. Using Russian-language social media posts, we analyze verbs of fear (to scare, to be afraid, etc.) and emotional attitude (to like, to love) with the semantic roles of experiencer, causator, and object. The study aims to compare self-assessed confidence of LLM Claude Sonnet 4.5 with expert assessments of the model's reasoning texts in argument classification in the healthcare domain. The experiment utilized a stratified set of 300 examples using chain-of-thoughts reasoning in Russian and four-step confidence scale. The results showed a moderate Spearman correlation between the expert and model assessments. A statistically significant relationship was found only between self-assessment of the model and the actual classification accuracy, while expert assessment of the linguistic characteristics of reasoning was unrelated to accuracy. It was concluded that explicit LLM reasoning is not directly related to its self-assessment of confidence and is separate from the decision-making process; reasoning may be an important functional part of the user interface, but not of the research.
Article Details
References
2. Zolotova G.A. Ocherk funkcional`nogo sintaksisa russkogo yazy`ka. M.: Nauka, 1973. 351 p.
3. Smirnov Ivan V., Larionov Daniil S., Nikitina Elena N., Kazachonok Grigory A. LLM for Semantic Role Labeling of Emotion Predicates in Russian // Supercomputing Frontiers and Innovations. 2025. Vol. 12, No. 3. P. 31–46. https://doi.org/10.14529/jsfi250303
4. Anthropic. The Claude 3 Model Family: Opus, Sonnet, Haiku. Technical Report. 2024.
5. Xiong M. et al. Can LLMs express their uncertainty? An empirical evaluation of confidence elicitation in LLMs // arXiv: 2306.13063. 2023. https://doi.org/10.48550/arXiv.2306.13063
6. Wei J. et al. Chain-of-thought prompting elicits reasoning in large language models //Advances in neural information processing systems. 2022. Vol. 35. P. 24824–24837.
7. Tian K.et al. Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback Proc. of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP). Singapore, 2023. P. 5433–5442. https://doi.org/10.18653/v1/2023.emnlp-main.330
8. Yang D., Tsai Y.H., Yamada M. On Verbalized Confidence Scores for LLMs // arXiv: 2412.14737. 2024. https://doi.org/10.48550/arXiv.2412.14737
9. Tanneru S.H., Agarwal C., Lakkaraju H. Quantifying Uncertainty in Natural Language Explanations of Large Language Models // Proc. of the 27th International Conference on Artificial Intelligence and Statistics. PMLR, 2024. Vol. 238. https://proceedings.mlr.press/v238/harsha-tanneru24a.html (accessed 05.05.2026)
10. Larionov D.S., Nikitina E.N., Smirnov I.V. Primenenie bol`shix yazy`kovy`x modelej dlya semanticheskoj klassifikacii argumentov pri e`motivny`x predikatax // Sistemy` vy`sokoj dostupnosti. 2026. Vol. 22, No. 1. P. 12−16. https://doi.org/10.18127/j20729472-202601-02
11. Nikitina E.N., Onipenko N.K. Semantika i pragmatika vyskazyvanij s emotivnymi predikatami // Sibirskij filologicheskij zhurnal. 2022. № 2. P. 271–285. https://doi.org/10.17223/18137083/79/19.
12. Nikitina E.N., Stankevich M.A., Larionov D.S. E`motivny`e glagoly` s otriczaniem: opy`t interpretacii danny`x avtomaticheskogo analiza tekstov // Medialingvistika. 2025. T. 12. №2. S. 174–198. https://doi.org/10.21638/spbu22.2025.201
13. Lyashevskaya O., Kashkin E. FrameBank: A Database of Russian Lexical Constructions // AIST 2015. Springer, 2015. P. 350–360.
14. Paducheva E.V. Modal'nost'. Materialy dlya proekta korpusnogo opisaniya russkoj grammatiki (rusgram.ru). Na pravah rukopisi. М. 2016.
15. Kuzmin G., Strepetov P., Stankevich M., Chudova N., Shelmanov A., Smirnov I. Exploring Large Language Models for Detecting Mental Disorders // Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 2025. P. 34535–34559. https://doi.org/10.18653/v1/2025.emnlp-main.1752
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.