DefinePK

Abstract

Idiomatic expressions pose a unique challenge in translation due to their culturally bound, non-literal meanings, and context-dependent usage. While artificial intelligence (AI) models, particularly neural machine translation (NMT) systems, have made significant advances in the field of language translation, the effective translation of idioms remains a complex task. This research explores the intersection of linguistics through a corpus-based study, aiming to evaluate the accuracy and cultural sensitivity of AI-generated idiom translations across multiple languages. Using a multilingual corpus that includes idioms from English, Urdu, Sindhi, and other selected languages, the study assesses the semantic fidelity and contextual appropriateness of AI translations. By identifying patterns of error and gaps in AI handling of idiomatic expressions, the research provides insights into the limitations of current AI models in translating culturally significant linguistic features. The findings of this study contribute to the improvement of AI translation systems, offering recommendations for refining algorithms to better handle idioms. Ultimately, this research aims to enhance cross-cultural communication and advance the integration of linguistics and AI in translation studies, ensuring greater accuracy and cultural relevance in AI-driven translations.