본문 바로가기
  • Home

ChatGPT prompt design for interpreter training content development: Focusing on text creation and editing

  • The Journal of Translation Studies
  • Abbr : JTS
  • 2024, 25(3), pp.123-150
  • DOI : 10.15749/jts.2024.25.3.005
  • Publisher : The Korean Association for Translation Studies
  • Research Area : Humanities > Interpretation and Translation Studies
  • Received : August 15, 2024
  • Accepted : September 14, 2024
  • Published : September 30, 2024

Jin Sil hee 1

1중앙대학교 통역번역연구소

Accredited

ABSTRACT

This paper provides an in-depth analysis of prompt design for large language models (LLMs), specifically ChatGPT, and its application in interpreter education. Positioned within the framework of Computer-Assisted Interpreter Training (CAIT), this study explores how ChatGPT can assist in generating and editing interpreter training contents. By integrating a comprehensive literature review with practical applications, the research provides concrete examples of ChatGPT prompts for two primary purposes: creating new source texts and editing existing content. For content generation, the study recommends using a CORE template, leveraging materials such as press releases, news articles, and other relevant documents to initiate a prompt series. Furthermore, it highlights the inclusion of difficulty elements in source texts, such as proverbs, idiomatic expressions, numbers, and other intricate features of text composition. This research underscores the transformative potential of AI-generated content (AIGC) in enhancing interpreter education, particularly by offering a scalable solution for creating diverse and challenging training materials. The study concludes by recommending further research to explore more sophisticated applications of LLMs in interpreter training programs. Despite some limitations, the findings emphasize that tools like ChatGPT represent a valuable resource for enriching interpreter education, indicating a promising future at the intersection of AI and pedagogy.

Citation status

* References for papers published after 2023 are currently being built.