EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Transliteration for Low-Resource Translation in the Age of Large Language Models

Aigerim Mansurova, Meruert Bekmukhamedova, Bekarys Baibolat, Nurassyl Beisenbek, Aliya Nugumanova, Beibit Abdikenov

Electronics · 10.07.2026 · 15 (14) · 3034

DOI: 10.3390/electronics15143034 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 10.07.2026. Verfasst von Aigerim Mansurova und 5 weiteren Autor:innen. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Science and Higher Education of the Republic of Kazakhstan.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neural machine translation (NMT) systems are widely used, but their performance remains strongly dependent on the availability of large-scale digital corpora, making translation for low-resource languages a persistent challenge. In parallel, large language models (LLMs) have recently emerged as a promising paradigm for multilingual text generation and translation; however, their behavior in low-resource settings remains largely underexplored. The challenge becomes even more acute for historical languages. Chagatai, a historical Turkic literary language of Central Asia with no native speakers, unstable orthography, and parallel data, represents an extreme case of such a condition. This study investigates whether transliteration significantly affects translation performance and how LLM-based and NMT-based systems compare under an extremely low-resource setting. To address these questions, we evaluated four source-text configurations (original Arabic script, expert manual transliteration, LLM-based transliteration, and rule-based Uroman transliteration) for translation into six target languages: Kazakh, English, Uzbek, Uyghur, Turkish, Russian, and Arabic. The results show that manual transliteration consistently yields the best translation performance, while noisy automatic romanization reduces these gains. For model comparison, GPT-4o was assessed alongside two fine-tuned NMT baselines, NLLB and TranslateGemma. The findings further show that LLM-based translation can be competitive with, and in some settings outperform, fine-tuned NMT systems, although this advantage comes with lower interpretability. Overall, these findings show that, for extremely low-resource historical languages written in non-Latin scripts, source-side representation is a decisive factor and may be as important as the choice of translation model itself.

Förderung

  • Ministry of Science and Higher Education of the Republic of Kazakhstan

Zitieren

APA

Mansurova, A., Bekmukhamedova, M., Baibolat, B., Beisenbek, N., Nugumanova, A., & Abdikenov, B. (2026). Transliteration for Low-Resource Translation in the Age of Large Language Models. Electronics, 15(14), 3034. https://doi.org/10.3390/electronics15143034

Vancouver

Mansurova A, Bekmukhamedova M, Baibolat B, Beisenbek N, Nugumanova A, Abdikenov B. Transliteration for Low-Resource Translation in the Age of Large Language Models. Electronics. 2026;15(14):3034. doi:10.3390/electronics15143034

BibTeX
@article{Mansurova2026,
  author = {Mansurova, Aigerim and Bekmukhamedova, Meruert and Baibolat, Bekarys and Beisenbek, Nurassyl and Nugumanova, Aliya and Abdikenov, Beibit},
  title = {Transliteration for Low-Resource Translation in the Age of Large Language Models},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {14},
  pages = {3034},
  doi = {10.3390/electronics15143034}
}
Kostenlos starten →