Generative AI for Translators and Language Specialists
About This Course
This 13-week course is designed for learners who have completed undergraduate studies in various disciplines and are interested in incorporating large language models (LLMs) into their translation workflow. The course covers the fundamentals of LLMs, prompting techniques, and prompt engineering, as well as integrate LLMs in several key phases of modern translation workflow such as pre-editing, post-editing, and quality evaluation. It also explores ethical concerns and productivity optimization in LLM-based machine translation. Through hands-on exercises, case studies, and final projects, learners will develop practical expertise in LLM-assisted translation workflows while critically assessing their advantages and limitations.
What You'll Learn
1) Explain key concepts of large language models (LLMs) and their relevance to modern translation workflows.
2) Apply effective prompting and prompt-engineering techniques to support translation, pre-editing, and post-editing tasks.
3) Use LLMs to perform and assist in translation and translation-quality evaluation.
4) Analyze LLM-generated translations to identify common errors, inconsistencies, and limitations.
5) Evaluate the reliability, quality, and ethical implications of LLM-assisted translation outputs.
6) Produce a project demonstrating practical, justified use of LLMs in translation practice.
Course Syllabus Outline:
a) Overview of the evolution of machine translation systems, with a focus on the development of large language models (LLMs) and their applications in translation;
b) Contrastive analysis of key linguistic differences between English and Chinese, and examination of the linguistic, cultural, and contextual challenges inherent in Chinese-English translation;
c) Interaction with LLMs through prompting, including core prompting strategies and principles of prompt engineering;
d) Integration of LLMs into the modern translation workflow, covering pre-editing, LLM-assisted translation, post-editing, and translation-quality evaluation;
e) Critical analysis of LLM-generated translations, including identification and categorisation of common error types;
f) Ethical considerations and data privacy issues in the use of generative artificial intelligence (GenAI) for translation;
g) Development and presentation of a final project, demonstrating the application of LLMs in a full translation workflow.
Entry Requirements
Admission Requirements
Applicants should hold a BA degree (at least an average grade of B or equivalent) preferably in language- or humanities-related disciplines. Candidates with other qualifications and experience may be considered on a case-by-case basis, subject to approval.
English Proficiency
Applicants whose native tongue and medium of university instruction are not English should submit their TOEFL or IELTS score. The following minimum score is required: 85 for the internet-based TOEFL test (with a minimum score of 22 for the writing section); or 237 for the computer-based TOEFL test; or IELTS score of 6.0.
Chinese Proficiency
Applicants should have a functional command of modern written Chinese sufficient for reading and understanding general and professional texts commonly encountered in Singapore. The minimum proficiency level is comparable to completion of secondary school Higher Chinese or equivalent. Native-like or advanced literary proficiency is NOT required.