A Study on the C-E Classic Sayings Translation Quality Assessment of LLM DeepSeek-R1 in Understanding Xi Jinping’s Educational Philosophy

Yang HE, Xue JIANG

Abstract


This study sampled 30 classic sayings contained in the Understanding Xi Jinping’s Educational Philosophy as the research objects. It takes the official translation version Understanding Xi Jinping’s Educational Philosophy which was commissioned to Beijing Foreign Studies University by the National Textbook Committee Office as the reference translation, and the translations generated by the domestically advanced Large Language Model (the following is abbreviated as LLM) DeepSeek-R1 as the objects for translation quality evaluation. The study adopts quantitative evaluation and qualitative analysis as research methods. Through the collaborative analysis of two indicators, namely COMET (Cross-lingual Optimized Metric for Evaluation of Translation) and TER (Translation Edit Rate), it quantitatively evaluates the performance of the evaluation objects in semantic fidelity, and the surface edit distance difference. At the same time, it qualitatively analyzes the cultural-loaded equivalence of the evaluation objects. The study finds that the LLM DeepSeek-R1 can translate well with accurate semantics and suitable contexts in the translation of classic sayings. However, when dealing with expressions with profound cultural connotations, there are still problems such as semantic deviations, insufficient transmission of cultural images, and a measurable surface edit distance from the reference translation. This study provides empirical evidence for enhancing the textual translation quality of the LLM DeepSeek-R1 in the field of classic sayings translation; it also serves as an important reference for future enhancements of the model in this field and the expansion of its application scenarios.

 


Keywords


LLM DeepSeek-R1; Translation quality assessment; Classic sayings translation; Understanding Xi Jinping’s Educational Philosophy

Full Text:

PDF

References


Devlin, J., Chang, M., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Retrieved from https://doi.org/10.48550/arXiv.1810.04805.

Feng, M. (2023). Translation for China’s Global Communication from the Perspective of Political Equivalence: A Case Study of Understanding Xi Jinping’s Educational Philosophy. Retrieved from https://doi.org/10.26962/d.cnki.gbjwu.2023.000150.

Fu, T. J. (2023). Knowledge Construction and Discourse Shaping in Translation Education: With Reference to the English Translation of The Excerpts from Xi Jinping’s Important Discourses on Education.Retrieved from https://doi.org/10.26914/c.cnkihy.2023.114163.

Hu, K. B. & Li, X. Q. (2023). The development of translation studies in the era of large language models: Challenges and prospects. Chinese Translators Journal, 44(06), 64-73+192.

Maike, Z., & Mrinmaya, S. (2025). COMET-poly: Machine Translation Metric Grounded in Other Candidates. Retrieved from https://doi.org/10.48550/arXiv.2508.18549.

Mo, C. (2017). On the English Translation of Characteristic Words in Chinese Classics from the Perspective of Equivalent Translation Theory. Journal of Suzhou University, 32(11), 62-65.

Mo, J. (2017). Discussion on the Application of Foreignization in News Translation-Han Shan Yi, Han Jie Fang Jun Nan as Example-Han Shan Yi, Han Jie Fang Jun Nan as Example. Journal of Beijing Institute of Graphic Communication, 25(05), 27-29.

Snover, M.G., & Makhoul, J. (2006). A Study of Translation Edit Rate with Targeted Human Annotation. Conference of the Association for Machine Translation in the Americas. Retrieved from https://api.semanticscholar.org/CorpusID:263887736.

Zhang, S. K., & Zhao,C. Y. (2025). A Study on the Appropriateness of Large Language Models in Literary Translation: A Comparative Analysis of the Translation Quality of Multiple Models for Border Town Based on Multi-Indicator Evaluation. Foreign Languages in China, 22(04), 85-95.




DOI: http://dx.doi.org/10.3968/13883

Refbacks

  • There are currently no refbacks.


Copyright (c) 2025 Cross-Cultural Communication

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Share us to:   


Remind

We are currently accepting submissions via email only.

The registration and online submission functions have been disabled.

Please send your manuscripts to ccc@cscanada.net,or  ccc@cscanada.org  for consideration. We look forward to receiving your work.

 

 Articles published in Cross-Cultural Communication are licensed under Creative Commons Attribution 4.0 (CC-BY).

 CROSS-CULTURAL COMMUNICATION Editorial Office

Address: 1055 Rue Lucien-L'Allier, Unit #772, Montreal, QC H3G 3C4, Canada.
Telephone: 1-514-558 6138 
Website: Http://www.cscanada.net; Http://www.cscanada.org 
E-mail:caooc@hotmail.com; office@cscanada.net

Copyright © Canadian Academy of Oriental and Occidental Culture