"Can make mistakes": Prompting ChatGPT to Enhance Literary MT output

Abstract

Operating at the intersection of generative artificial intelligence, machine translation, and literary translation, this paper examines to what extent prompt-driven post-editing can enhance the quality of machine-translated lit- erary texts. We assess how different types of instruction influence post-editing performance, particularly focusing on literary nuances and author-specific styles. Situated within posthumanist translation theory, which often challenges traditional notions of human intervention in translation processes, the study explores the practical implementation of generative artificial intelligence in multi-lingual workflows. While the findings suggest that prompted post-editing can improve translation output to some extent, its effectiveness varies, especially in literary contexts. This highlights the need for a critical review of prompt engineering approaches and emphasizes the importance of further research to navigate the complexities of integrating AI into creative translation workflows effectively.