Is machine translation the greatest thing since sliced bread? Human and machine translation of food-related idioms
DOI:
https://doi.org/10.5007/2175-7968.2026.e110606Palavras-chave:
food idioms, machine translation, large language models, neural machine translation, human translationResumo
This study examines the translation of food-related idiomatic expressions (IEs) from English into Brazilian Portuguese, by comparing the performance of trainee translators and machine translation (MT) systems operating under two distinct paradigms: neural machine translation (NMT) and large language models (LLMs). Contexts containing IEs are extracted from the Corpus of Contemporary American English and translated by undergraduates, and by four AI-based systems: Google Translate and DeepL (NMT-based), and ChatGPT-4 and Gemini 2.5 Pro (LLM-based). All translations are qualitatively evaluated for idiomaticity and conventionality using corpus evidence and classified into four categories: idiomatic and conventional, non-idiomatic but conventional, idiomatic but non-conventional, and inaccurate. When human translation (HT) and MT are compared overall, HT produces a higher proportion of idiomatic and conventional equivalents. While both approaches show similar proportions of non-idiomatic but conventional and idiomatic but non-conventional renderings, inaccurate translations are almost twice as frequent in MT. When the performance of HT is compared with that of the LLM-based systems, the results indicate that the latter, largely driven by Gemini’s performance, slightly outperform students in producing appropriate translations. A more fine-grained analysis, however, reveals important differences within MT: NMT systems perform markedly worse, generating few idiomatic and conventional equivalents and a high proportion of inaccurate translations. Statistical analyses confirmed a significant association between translator type (human, NMT, or LLM) and translation quality (idiomatic and conventional, non-idiomatic but conventional, idiomatic but non-conventional, or inaccurate) across the four evaluation categories. This association is driven primarily by the poorer performance of NMT systems, whereas no significant difference is found between human translators and LLMs. These results highlight the importance of integrating human expertise with advanced MT tools in idiom-rich translation tasks. They also highlight the need for translator training that emphasizes idiomaticity, conventional usage, and critical evaluation of machine-generated output, preparing translators to work effectively in increasingly hybrid translation environments.
Referências
Alisoy, H. (2025). A Taxonomic Approach to Structural and Semantic Dimensions in English Phraseology. Porta Universorum, 1(4), 72–79. https://doi.org/10.69760/portuni.0104007
Angelone, E. (2026). Generative AI as a facilitator of deliberate practice in translator training. In J. C. Penet, J. Moorkens & M. Yamada (Eds.), Teaching translation in the age of generative AI: New paradigm, new learning? (pp. 27–47). Language Science Press.
Batista, J. S., Moreira de Oliveira, I., Tagnin, S. E. O., Teixeira, E. D., & Rebechi, R. R. (2026). Classificadores semânticos na identificação de equivalentes interlinguais para expressões idiomáticas: descascando esse abacaxi. Letras & Letras, 42(suppl.), 1–26. https://doi.org/10.14393/LL63-v42S-20226-0
Baziotis, C., Mathur, P., & Hasler, Eva. (2023). Automatic Evaluation and Analysis of Idioms in Neural Machine Translation. In A. Vlachos & I. Augenstein (Eds.), Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (pp. 3682–3700). Association for Computational Linguistics.
Biber, D., Johansson, S., Leech, G., Conrad, S., & Finegan, E. (1999). Longman Grammar of Spoken and Written English. Pearson Education Limited.
Cambridge University Press. (n.d.). Bring home the bacon. In Cambridge Dictionary. https://dictionary.cambridge.org/dictionary/english/bring-home-the-bacon
Castaldo, A., & Monti, J. (2024). Prompting Large Language Models for Idiomatic Translation. In B. Vanroy, M.-A. Lefer, L. Macken & P. Ruffo (Eds.), Proceedings of the 1st Workshop on Creative-text Translation and Technology (pp. 32–39). European Association for Machine Translation.
Davies, M. (2008-). The Corpus of Contemporary American English (COCA). https://www.english-corpora.org/coca/
Davies, M. (2016-). Corpus do Português: Web/Dialects. http://www.corpusdoportugues.org/web-dial/
Davies, M. (2012-2019). Corpus do Português: NOW. https://www.corpusdoportugues.org/now/
Donthi, S., Spencer, M., Patel, O. B., Doh, J. Y., Rodan, E., Zhu, K., & O’Brien, S. (2025). Improving LLM abilities in idiomatic translation. In H. Hettiarachchi, T. Ranasinghe, P. Rayson, R. Mitkov, M. Gaber, D. Premasiri, F. A. Tan & U. Lasitha (Eds.), Proceedings of the First Workshop on Language Models for Low-Resource Languages (pp. 175–181).
Erman, B., & Warren, B. (2000). The idiom principle and the open choice principle. Text, 20(1), 29–62. https://doi.org/10.1515/text.1.2000.20.1.29
Hutchins, W. J. (1986). Machine translation: Past, present, future. Ellis Horwood.
Jiang, R., Liu, X., Liu S., Wang, Y., Zhang, M., Tao, S., Wei, D., & Zhang, M. (2026). DeReA: Improving Idiom Translation with Detect-Retrieve-Arbitrate Reasoning. In M. Liakata, V. Moreira, J. Zhang & D. Jurgens (Eds.), Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics [Long Papers] (pp. 6603–6621). Association for Computational Linguistics. https://doi.org/10.18653/v1/2026.acl-long.299
Jiao, W., Wang, W., Huang, J., Wang, X., Shi, S., & Tu, Z. (2023). Is ChatGPT a Good Translator? Yes with GPT-4 as the Engine. arXiv:2301.08745v4, 1–10. https://doi.org/10.48550/arXiv.2301.08745
Jibreel, I. (2024). Translation Quality of Artificial Intelligence and Machine Translation Vs. Human Translation Utilizing MTPE Skills (An Empirical Study on Allusion Translation). Journal of Social Studies, 30(3), 46–72, https://doi.org/10.20428/jss.v30i3.2545
Joshi, P., Santy, S., Budhiraja, A., Bali, K., & Choudhury, M. (2020). The state and fate of linguistic diversity and inclusion in the NLP world. In D. Jurafsky, J. Chai, N. Schluter & J. Tetreault (Eds.), Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (pp. 6282–6293). Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.acl-main.560
Jurafsky, D., & Martin, J. H. (2026). Speech and language processing: An introduction to natural language processing, computational linguistics, and speech recognition (3rd ed). [Online manuscript]. https://web.stanford.edu/~jurafsky/slp3/
Mammadova, I. (2025). Machine Translation vs. Human Translation: A Linguistic Analysis. Porta Universorum, 1(1), 26-31. https://doi.org/10.69760/vhrq8s76
Massey, G., & Ehrensberger-Dow, M. (2026). Translation competence in the age of generative AI: Debates, dilemmas, directions. In J. C. Penet, J. Moorkens & M. Yamada (Eds.), Teaching translation in the age of generative AI: New paradigm, new learning? (pp. 3–25). Language Science Press.
McEnery, T., & Hardie, A. (2012). Corpus Linguistics. Cambridge University Press.
Montesinos, L. S., Buján, S., Bardanca D., & Gamallo, P. (2026). Improving Machine Translation of Idioms: A Spanish–Galician Parallel Dataset and Synthetic Augmentation Approach. In M. Souza, I. de-Dios-Flores, D. Santos, L. Freitas, J. W. C. Souza & E. Ribeiro (Eds.), Proceedings of the 17th International Conference on Computational Processing of Portuguese (Vol. 1; pp. 980–987). Association for Computational Linguistics.
Moreira de Oliveira, I. (2022). Expressões idiomáticas com a temática alimentação: uma proposta de glossário Português – Inglês. [Dissertação de Mestrado]. Universidade de Brasília.
Moreira de Oliveira, I., & Teixeira, E. (2025). Expressões idiomáticas com a temática alimentação: um estudo contrastivo português inglês. In A. Husky, S. Mejri & C. S. N. Salvador (Orgs.), A Fraseologia: O paradoxo do universal e do idiomático (pp. 195–236). Pontes Editores.
Musaad, M. M. A. M., & Al Towity, D. A. A. (2023). Translation Evaluation of Three Machine Translation Systems, with Special References to Idiomatic Expressions. Humanities and Educational Sciences Journal, 29, 678–708. https://doi.org/10.55074/hesj.vi29.700
Naveen, P., & Trojovský, P. (2024). Overview and challenges of machine translation for contextually appropriate translations. iScience, 27(10), 1–25. https://doi.org/10.1016/j.isci.2024.110878
Oliveira, M. L. S., Santos, A. A. O., & Andrade, L. J. S. (2026). Bridging Cultural Gaps in Automated Translation of Brazilian Expressions: A Study on Cultural Adaptation. In M. Souza, I. de-Dios-Flores, D. Santos, L. Freitas, J. W. C. Souza & E. Ribeiro (Eds.), Proceedings of the 17th International Conference on Computational Processing of Portuguese (Vol. 2; pp. 220–227). Association for Computational Linguistics.
Pang, J., Ye, F., Wong, D. F., Yu, D., Shi, S., Tu, Z., & Wang, L. (2025). Salute the classic: Revisiting challenges of machine translation in the age of large language models. Transactions of the Association for Computational Linguistics, 13, 73–95. https://doi.org/10.48550/arXiv.2401.08350
Penet, J. C., Moorkens, J., & Yamada, M. (Eds.). (2026). Teaching translation in the age of generative AI: New paradigm, new learning? Language Science Press. https://doi.org/10.5281/zenodo.17580856
Puppel, M., & Borg, C. (2025). Evaluating ChatGPT’s performance in creative text translation for communication: A case study from English into German. Media and Intercultural Communication, 3(1), 1–27. https://doi.org/10.22034/mic.2024.480506.1023
Ramisch, C., Cordeiro, S., Zilio, L., Idiart, M., Villavicencio, A. (2016) How Naked is the Naked Truth? A Multilingual Lexicon of Nominal Compound Compositionality. In K. Erk & N. Smith (Eds.), Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Vol. 2; pp. 156–161). Association for Computational Linguistics. https://doi.org/10.18653/v1/P16-2026
Rebechi, R., & Alves, D. A. (2026). A Comparative Study of LLMs for Idiomatic Expression Identification and Translation Equivalent Extraction in an English–Portuguese Subtitle Corpus. In M. Escribe, A. P. Izquierdo, C. Orǎsan, T. Ranasinghe, G. C. Pastor, M. Tadic & R. Mitkov (Eds.), Proceedings of the 3rd International Conference New Trends in Translation and Interpreting Technology 2026 (Vol. 1; pp. 54–62). Incoma.
Rebechi, R. R., Marcon, N. O., & Faller, G. A. (2025). Tradução automática e tradução humana de expressões multipalavras: Descascando esse abacaxi. ReVEL, 23(44), 346–380.
Rebechi, R. R., Marcon, N. O., Faller, G. A., & Villavicencio, A. (2024). Machine Translation of Multiword Expressions: Double Dutch or Crystal Clear? In C. Orǎsan, T. Ranasinghe, G. C. Pastor, R. Mitkov, M. Kunilovskaya, V. Sosoni & M. Escribe (Eds.), Proceedings of the Conference New Trends in Translation and Technology 2024: Translation in the AI age (pp. 116–137). Incoma.
Rebechi, R. R., & Trindade, E. (2021). Traduzir metáfora não é mamão com açúcar: A busca por equivalentes de botanomorfismos. Polissema, 21, 110–132.
Rizki, K., & Masykuroh, Q. (2025). Evaluating ChatGPT's Translation of Harry Potter: A Qualitative Study of Translation Techniques, Accuracy, and Acceptability. Journal of English Language Teaching and Literature, 6(1), 181–192.
Rothwell, A., Moorkens, J., Fernández-Parra, M., Drugan, J., & Austermuehl, F. (2023). Translation tools and technologies. Routledge.
Schmid, H.-J. (2020). The Dynamics of the Linguistic System: Usage, Conventionalization, and Entrenchment. Oxford University Press.
Sinclair, J. (1991). Corpus, Concordance, Collocation. Oxford University Press.
Tagnin, S. E. O. (2013). O jeito que a gente diz: Combinações consagradas em inglês e português. Disal.
Xatara, C., Riva, H. C., & Rios, T. H. C. (2001). As dificuldades na tradução de idiomatismos. Cadernos de Tradução, 2(8), 183–194.
Downloads
Publicado
Como Citar
Edição
Seção
Licença
Copyright (c) 2026 Cadernos de Tradução
Este trabalho está licenciado sob uma licença Creative Commons Attribution 4.0 International License.
Autores têm autorização para assumir contratos adicionais separadamente, para distribuição não exclusiva da versão do trabalho publicada nesta revista (ex.: publicar em repositório institucional ou como capítulo de livro, com reconhecimento de autoria e publicação inicial nesta revista).
