"The Infinite" by Leopardi
Morphological Collocation in Portuguese Translations Generated by Artificial Intelligences
DOI:
https://doi.org/10.35699/2317-2096.2026.62984Keywords:
literary translation, poetry translation, Artificial IntelligenceAbstract
This article presents a comparative study of the translations of the poem “The Infinite”, by Giacomo Leopardi, generated by artificial intelligences (ChatGPT versions 5.0 paid and 4.0 free, and Gemini versions 2.5 pro paid and free), focusing on morphological structures. The research stems from the growing presence of AIs in literary translation and the lack of empirical studies on the quality of these translations. The methodology combines three procedures: the creation of a guiding agent to assist in the development of prompts, morphological tagging with the Stanza software, and n-gram analysis using the ROUGE technique. The hypothesis is that AI-generated translations may show significant variations in morphological structures. The results indicate that, quantitatively, AIs maintain much of the source text’s morphological structures in the translations, and that many issues stem from the improper formulation of prompts. Qualitatively, it was observed that the AIs were able to preserve the Leopardi style, especially in the placement of words.
Downloads
References
GOMATHI, R. D. et al. AI-Driven Literary Analysis: Exploring the Impact of Artificial Intelligence on Text Interpretation and Criticism. Journal of Machine and Computing, [s. l.], v. 5, n. 2, p. 1124-1139, abr. 2025. DOI: https://doi.org/10.53759/7669/jmc202505089. Disponível em: doi.org/10.53759/7669%2Fjmc202505089?sid=semanticscholar. Acesso em: 5 jul. 2025.
GOMES, Dennis dos Santos. Inteligência Artificial: conceitos e aplicações. Revista Olhar Científico, Ariquemes, v. 1, n. 2, p. 234-246, 2010.
GUERINI, Andréia; COSTA, Walter. Colocação e qualidade na poesia traduzida. Tradução em revista, Rio de Janeiro, v. 2024, n. 36, 2006. DOI: 10.17771/PUCRio.TradRev.9358. Disponível em: https://web.archive.org/web/20220127001720id_/https://www.maxwell.vrac.puc-rio.br/9358/9358.PDF. Acesso em: 8 maio 2025.
GUERINI, Andréia. “L'infinito”: tensão entre teoria e prática na tradução de Haroldo de Campos. Cadernos de Tradução, Florianópolis, v. 26, n. 2, p. 105-114, 2000. DOI: https://doi.org/10.5007/%25x. Disponível em: https://periodicos.ufsc.br/index.php/traducao/article/view/5684/5184. Acesso em: 14 jul. 2025.
LEOPARDI, Giacomo. Tutte le poesie, tutte le prose e lo Zibaldone. Edizione integrale diretta da Lucio Felici e Emanuele Trevi. Roma: Newton Compton, 2010.
LI, Qi. Bridging Languages: The Potential and Limitations of AI in Literary Translation. A Case Study of the English Translation of A Pair of Peacocks Southeast Fly. Advances in Humanities Research, [s. l.], v. 8, n. 1, p. 1-7, ago. 2024. DOI: 10.54254/2753-7080/8/2024091. Disponível em: https://www.ewadirect.com/journal/ahr/article/view/15283. Acesso em: 13 jun. 2025.
SPOTURNO, María Laura. Traducción literaria e inteligencia artificial: consideraciones para la formación universitaria. Cadernos de Tradução, Florianópolis, v. 44, n. 1, p. 1-26, 2024. DOI: 10.5007/2175-7968.2024.e100602. Disponível em: https://periodicos.ufsc.br/index.php/traducao/article/view/100602. Acesso em: 19 mar. 2025.
TAGAWA, Thaís Yassue Domingos; AGUIAR, Isabella Correia; VILLANI, Leonardo. Eficácia e impacto na experiência do cliente: um estudo sobre os modelos generativos ChatGPT e Gemini. Revista Processando o Saber, São Paulo, v. 17, n. 1, p. 197-211, jun. 2025. DOI: https://doi.org/10.5281/zenodo.15486309. Disponível em: https://www.fatecpg.edu.br/revista/index.php/ps/article/view/396/269. Acesso em: 15 abr. 2025.
WANG, Qiufen. Evaluating Uighur Literary Translation: A Comparative Study of ChatGPT, Google Translate, and Bing Translator. PLoS One, [s. l.], v. 20, n. 10, [n. p], out. 2025. DOI: https://doi.org/10.1371/journal.pone.0335261. Acesso em: 5 dez. 2025.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Ingrid Bignardi, Andréia Guerini (Autor)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution Non-Commercial No Derivatives License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
How to Cite
