Accurate translation of terminology and adaptation to in-context information is a pillar to high quality translation. Recently, there is a remarkable interest towards the use and the evaluation of Large Language Models (LLMs) particularly for Machine Translation tasks. Nevertheless, despite their recent advancement and ability to understand and generate human-like language, these LLMs are still far from perfect, especially in domain-specific scenarios, and need to be thoroughly investigated. This is particularly evident in automatically translating legal terminology from Arabic into English and French, where, beyond the inherent complexities of legal language and specialised translations, technical limitations of LLMs further hinder accurate generation of text. In this paper, we present a preliminary evaluation of two evolving LLMs, namely GPT-4 Generative Pre-trained Transformer and Gemini, as legal translators of Arabic legislatives to test their accuracy and the extent to which they care for context and terminology across two language pairs (AR→EN / AR→FR). The study targets the evaluation of Zero-Shot prompting for in-context and out-of-context scenarios of both models relying on a gold standard dataset, verified by professional translators who are also experts in the field. We evaluate the results applying the Multidimensional Quality Metrics to classify translation errors. Moreover, we also evaluate the general LLMs outputs to verify their correctness, consistency, and completeness. In general, our results show that the models are far from perfect and recall for more fine-tuning efforts using specialised terminological data in the legal domain from Arabic into English and French.

In the translation process, terminological resources are used to solve translation problems, so information on terminological equivalence is crucial to make the most appropriate choices in terms of translation equivalence. In the context of Machine translation, indeed, neural models have improved the state-of-the-art in Machine Translation considerably in recent years. However, they still underperform in domain-specific fields and in under-resourced languages. This is particularly evident in translating legal terminology for Arabic, where current Machine Translation outputs do not adhere to the contextual, linguistic, cultural, and terminological constraints posed by translating legal terms in Arabic. In this paper, we conduct a comparative qualitative evaluation and comprehensive error analysis on legal terminology translation in Phrase-Based Statistical Machine Translation and Neural Machine Translation in two translation language pairs: Arabic-English and Arabic-French. We propose an error typology taking the legal terminology translation from Arabic into account. We demonstrate our findings, highlighting the strengths and weaknesses of both approaches in the area of legal terminology translation for Arabic. We also introduce a multilingual gold standard dataset that we developed using our Arabic legal corpus. This dataset serves as a reliable benchmark and/or reference during the evaluation process to decide the degree of adequacy and fluency of the Phrase-Based Statistical Machine Translation and Neural Machine Translation systems.