Interpretation of figurative language in text by linguistic neural networks Alice YandexGPT 5 Pro, GPT-4o DUM-E, DeepSeek Rico and GigaChat
https://doi.org/10.37493/2409-1030.2025.3.20
Abstract
Introduction. The purpose of the research was to study the abilities of the neural networks Alice YandexGPT 5 Pro, GPT-4o DUM-E, DeepSeek Rico, GigaChat to adequately interpret literary texts. The following possibilities of these linguistic neural networks were studied: identification of a stylistic device in the text; application of the language of meta-description; recognition and characterization of precedent phenomena – historical characters, events, phenomena; recognition and interpretation of metaphor and hyperbole; interpretation of semantically complex expressions in context.
Materials and methods. A fragment of the first chapter of the novel The Golden Calf by I. Ilf and E. Petrov was chosen as the research material. The method of macrostructural analysis, the method of interpretative analysis, and procedural-semantic analysis were used as the main ones.
Analysis. In the course of the study, rhematic elements were identified in the responses-interpretations of neural networks, summarized in a single table and subjected to semantic and comparative analysis, which revealed both successful interpretations and negative results.
Results. It is shown that the studied language neural networks als and methods. A fragment of the first chapter of the novel The Golden Calf by I. Ilf and E. Petrov was chosen as the research material. The method of macrostructural analysis, the method of interpretative analysis, and procedural-semantic analysis were used as the main ones. Analysis. In the course of the study, rhematic elements were identified in the responses-interpretations of neural networks, summarized in a single table and subjected to semantic and comparative analysis, which revealed both successful interpretations and negative results. Results. It is shown that the studied language neural networks.
About the Authors
S. V. Gusarenko
North-Caucasus Federal University
Russian Federation
Russian Federation
Sergey V. Gusarenko - Dr. Sc. (Philology), Professor
1, Pushkina St., Stavropol, 355017
M. K. Gusarenko
North-Caucasus Federal University
Russian Federation
Russian Federation
Marina K. Gusarenko - Cand. Sc. (Philology), Associate Professor
1, Pushkina St., Stavropol, 355017
References
1. Alice YandexGPT 5 Pro. Available from: https://alice.yandex.ru/ [Accessed 25 June 2025]. (In Russ.).
2. Arutyunova ND. The sentence and its meaning: logical-semantic problems. Moskow: Nauka; 1976. 383 p. (In Russ.).
3. Grice GP. Logics and conversation. New in foreign linguistics: Linguistic pragmatics. Moscow: Progress; 1985;(XVI):217-237. (In Russ.).
4. Galperin IR. Text as an object of linguistic research. Moscow: Editorial URSS; 2004. 144 p. (In Russ.).
5. Deik TA. van, Kinch V. Strategies for understanding a coherent text. New in foreign linguistics: Cognitive aspects of language. 1988;(XXIII):153-211. (In Russ.).
6. Paducheva EV. Utterance and its correlation with reality (referential aspects of the semantics of pronouns). 4th ed. Moscow: Editorial; 2004. 288 p. (In Russ.).
7. Bazin H. Are Large Language Models capable of Gricean conversational implicatures? Available from: https://sopha2025.sciencesconf.org/browse/author?authorid=1323099 [Accessed 25 June 2025].
8. DeepSeek Rico. Available from: https://web.telegram.org/a/#6433748390 [Accessed 25 June 2025].
9. Testing the Ability of Language Models to Interpret Figurative Language / L. Emmy, C. Chenxuan, Zh. Kenneth, N. Graham. Available from: https://arxiv.org/pdf/2204.12632v1 [Accessed 25 July 2025].
10. GigaChat. Available from: https://web.telegram.org/a/#6218783903 [Accessed 25 June 2025].
11. GPT-4o DUM-E. Available from: https://web.telegram.org/a/#6010034370 [Accessed 25 June 2025].
12. Gromenko E., Kalacheva D., Klokova K., Krongauz M., Moroz O., Shulginov V., Yudina T. Cultural Evaluation of LLMs in Russian: Catchphrases and Cultural Types. Available from: https://dialogue-conf.org/wp-content/uploads/2025/06/GromenkoEetal.029.pdf [Accessed 25 June 2025].
13. Gu Y., Fu Y., Pyatkin V., Magnusson I., Mishra B.D., Clark P. Just-DREAM-about-it: Figurative Language Understanding with DREAM-FLUTE. URL: https://arxiv.org/pdf/2210.16407 [Accessed 25 June 2025].
14. Kotov A.A., Filatov A.A., Nosovets Z.A., Arinkin N.A. The development of a natural language reasoning system for a companion robot. URL: https://dialogue-conf.org/wp-content/uploads/2025/06/KotovAAetal.077.pdf [Accessed 25 June 2025].
15. Shulginov V., Şimşek H.B., Kudriashov S., Shevela S.A., Randavtsova R. Evaluating the Pragmatic Competence of Large Language Models in Detecting Mitigated and Unmitigated Types of Disagreement. Available from: https://dialogueconf.org/wp-content/uploads/2025/06/ShulginovVetal.037.pdf [Accessed 25 June 2025].
Review
For citations:
Gusarenko S.V., Gusarenko M.K. Interpretation of figurative language in text by linguistic neural networks Alice YandexGPT 5 Pro, GPT-4o DUM-E, DeepSeek Rico and GigaChat. Humanities and law research. 2025;12(3):513-523. (In Russ.) https://doi.org/10.37493/2409-1030.2025.3.20
Views:
53
Email this article 