Current LLMs and English syntax terminology translation

Read full text Summarise with VIRAT
Authors: Mohammed Q. Shormani, Yehia A. AlSohbani
Cited by: 0
References: 55

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

We aim to examine the extent to which Large Language Models (LLMs), specifically ChatGPT-5 can translate English syntax terminology into Arabic. We collected 44 terms from generative syntax previous works, including books and journal articles, as well as from our experience in the field. These terms were translated by humans, and then by ChatGPT-5. We then analyzed and compared both translations. We used the analytical and comparative approach amalgamated with micro approach in our analysis. Findings unveil that ChatGPT still cannot translate the core syntax properties embedded in the terms under study involving several syntactic (and semantic) challenges: only 25% of ChatGPT translations were accurate, while 38.6% were inaccurate, and 36.4.% were partially correct, which we consider appropriate. Based on these findings, a set of actionable strategies were proposed, the most notable of which is finetuning LLMs on syntactic terminology in both languages to better LLMs’ working mechanism for accurate or at least appropriate translation.

Fields of Study & Subjects

Natural Language Processing Techniques Translation Studies and Practices Topic Modeling

Publication details

Journal / source
Discover Artificial Intelligence
Volume
6
Issue
1
Publication date
9 November 2026
Language
EN