magazinelogo

Journal of Humanities, Arts and Social Science

ISSN Online: 2576-0548 ISSN Print: 2576-0556 CODEN: JHASAY
Frequency: monthly Email: jhass@hillpublisher.com
Total View: 6569035 Downloads: 1928240 Citations: 437 (From Dimensions)
ArticleOpen Access http://dx.doi.org/10.26855/jhass.2026.06.015

GPT-4o vs. ERNIE 5.1: Strategic Differentiation and Cultural Discount in Chinese Menu English Translation via Cultural Prompts

Mei Luo

School of Foreign Languages, Guangxi University of Chinese Medicine, Nanning 530001, Guangxi, China.

*Corresponding author: Mei Luo

This research is supported by the 2025 Annual Project of Guangxi Philosophy and Social Sciences Research (Grant No. 25GJF053): “Research on Image Construction and Strategy Optimization for the Dissemination of Zhuang and Yao Medical Culture in the ASEAN Region.”
Published: June 30, 2026

Abstract

When large language models (LLMs) translate culturally loaded Chinese dish names, their differing strategies can produce distinct patterns of cultural discount—an issue that has received little systematic attention. This paper compares how GPT-4o and ERNIE 5.1 handle three Guangxi specialties under an identical cultural prompt, and traces the losses that emerge. The analysis is organized around three dimensions: cultural information encoding, functional orientation conversion, and metaphorical cognitive transmission. The findings show that both models use the “primary translation + annotation” format, but differ in their annotations. GPT-4o assimilates source culture into target-language frameworks, leading to encoding and functional losses; ERNIE 5.1 supplies richer cultural detail but with occasional factual inaccuracies, so losses remain mostly at the encoding level. Notably, metaphorical transmission remains a shared weakness, with neither model able to bridge the experiential gap. Drawing on these findings, we propose a three-stage “model complementarity + human dominance” workflow in which the translator acts as an evaluator and integrator of outputs from multiple models. This reorientation, we argue, provides a practical pathway for translating culinary culture when LLMs are involved.

Keyword

LLMs; cultural prompts; Chinese menu English translation; strategic differentiation; cultural discount

References

Amenador, K. B., & Wang, Z. (2022). The translation of culture-specific items (CSIs) in Chinese-English food menu corpus: A study of strategies and factors. SAGE Open, 12(2), 1-17.
https://doi.org/10.1177/21582440221096649

Amenador, K. B., & Wang, Z. (2023). The image of China as a destination for tourist: Translation strategies of culture-specific items in the Chinese-English food menus. SAGE Open, 13(3), 1-14. 

https://doi.org/10.1177/21582440231196656

Cao, Y., Kementchedjhieva, Y., Cui, R., Karamolegkou, A., Zhou, L., & Dare, M., et al. (2024). Cultural adaptation of recipes. Transactions of the Association for Computational Linguistics, 12, 80-99. 

https://doi.org/10.1162/tacl_a_00634

Hoskins, C., & Mirus, R. (1988). Reasons for the US dominance of the international trade in television programmes. Me-dia, Culture & Society, 10(4), 499-515.

Hu, G. (2004). An approach to translation as adaptation and selection. Hubei Education Press.

Liu, M., Jian, X., & Xu, J. (2026). Translation of culture-loaded words and cross-cultural communication in AI large language models—A case study of “The Return of the Condor Heroes”. Modern Linguistics, 14(4), 851-859. 

https://doi.org/10.12677/ml.2026.144363

Nida, E. A. (1993). Toward a science of translating (2nd ed.). Brill.

Venuti, L. (1995). The translator’s invisibility: A history of translation. Routledge.

Yu, G. (2025). Research on the “alignment mechanism” in the new ecology of cross-cultural communication: An explo-ration of communication paths based on cultural discount dissolution. International Communications, (8), 58-61.

https://doi.org/10.26948/j.cnki.dwcb.2025.08.016

Zhang, Z., He, X., Iyer, V., & Birch, A. (2024). Cultural adaptation of menus: A fine-grained approach. In Proceedings of the Ninth Conference on Machine Translation (WMT 2024) (pp. 1258-1271). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.wmt-1.120

Zhao, Y., Zhang, H., & Yang, Y. (2024). Comparative study on the translation quality of large language models—Taking the translation of “Fan Hua” as an example. Technology Enhanced Foreign Language Education, (4), 60-66, 109.

https://doi.org/10.20139/j.issn.1001-5795.20240409

Zhu, H., Ang, L. H., & Mansor, N. S. (2024). Are mouse noodles actually made from mice? Touring street food name translations. Humanities and Social Sciences Communications, 11(1), 1-10. 

https://doi.org/10.1057/s41599-024-03041-1

Copyright

© 2026 by the author(s).
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial-NoDerivatives (CC BY-NC-ND) license, which permits non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited and is not modified or adapted.
https://creativecommons.org/licenses/by-nc-nd/4.0/

How to cite this paper

GPT-4o vs. ERNIE 5.1: Strategic Differentiation and Cultural Discount in Chinese Menu English Translation via Cultural Prompts

How to cite this paper: Mei Luo. (2026) GPT-4o vs. ERNIE 5.1: Strategic Differentiation and Cultural Discount in Chinese Menu English Translation via Cultural Prompts. Journal of Humanities, Arts and Social Science10(6), 711-717.

DOI: http://dx.doi.org/10.26855/jhass.2026.06.015