Towards Engineering Design Principles for Role-Based AI Translation Systems

Pavel Petrov, Vassil Milev, Filipp Safronov, Vladimir Filatov

Abstract


This paper explores the engineering principles of role-based AI translation systems. A configurable experimental framework was developed based on a review of existing multi-agent translation architectures to evaluate role allocation, model selection, glossary integration, and review strategies. Controlled experiments have identified the key factors that influence translation quality, and have formulated practical design principles for the development of efficient, domain-specific Role-Based AI translation systems.

Keywords


role-based machine translation; multi-agent systems; AI translation engineering; terminology integration; local language models; computational efficiency

Full Text:

PDF

References


V. Briva-Iglesias, Are AI Agents the New Machine Translation Frontier? Challenges and Opportunities of Single- and Multi-Agent Systems for Multilingual Digital Communication, Proceedings of Machine Translation Summit XX, vol. 1 (2025), 365-377.

A.R. Ibrahimzada, et al., ReCodeAgent: A Multi-Agent Workflow for Language-Agnostic Translation and Validation of Large-Scale Repositories, arXiv preprint arXiv:2604.07341 (2026).

M. Li, K. Zhang, A Multi-Agent System Based on HNC for Domain-Specific Machine Translation, Scientific Reports, vol. 15, no. 1 (2025), 20820.

A. Lommel, et al., The Multi-Range Theory of Translation Quality Measurement: MQM Scoring Models and Statistical Quality Control, arXiv preprint arXiv:2405.16969 (2024).

V. Milev, F. Safronov, M. Racheva, A Model for Customizing AI Assistants for Specialized Translation for Educational Purposes, Digital Educational Technologies (2026), 55-67, https://doi.org/10.54664/NYFS9761.

W. Minghao, et al., (Perhaps) Beyond Human Translation: Harnessing Multi-Agent Collaboration for Translating Ultra-Long Literary Texts, Transactions of the Association for Computational Linguistics, vol. 13 (2025), 901-922.

A. Nuraeni, N.S. Saptaningsih, D. Adika, Role-Based Prompting for Translation: A Study on Language Style and Translation Technique of Airline Slogan Translation Generated by ChatGPT, LiNGUA, vol. 21, no. 1 (2026), 49.

K. Papineni, et al., Bleu: A Method for Automatic Evaluation of Machine Translation, Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (2002), 311-318.

M. Popović, chrF++: Words Helping Character n-Grams, Proceedings of the Second Conference on Machine Translation (2017), 612-618.

R. Rei, et al., COMET: A Neural Framework for MT Evaluation, Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) (2020), 2685-2702.

R. Sajja, et al., Artificial Intelligence-Enabled Intelligent Assistant for Personalized and Adaptive Learning in Higher Education, Information, vol. 15, no. 10 (2024), 596.

R. Sharma, M. Mehta. Small language models for agentic systems: A survey of architectures, capabilities, and deployment trade-offs. arXiv preprint arXiv:2510.03847, (2025).

B. Tianchi, et al., Multi-Agent Learning for Neural Machine Translation, Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) (2019), 856-865.

G. Wang, J. Hu, S. Ali, MAATS: A Multi-Agent Automated Translation System Based on MQM Evaluation, arXiv preprint arXiv:2505.14848 (2025).

S. Zhang, et al., HiMATE: A Hierarchical Multi-Agent Framework for Machine Translation Evaluation, Findings of the Association for Computational Linguistics: EMNLP (2025), 11121-11145.




DOI: http://dx.doi.org/10.5281/zenodo.21806584

Refbacks

  • There are currently no refbacks.
We use cookies.