Metaphors in Literary Post-Editing: Opening Pandora's Box?
A new paper explores the challenges literary post-editors face when translating metaphors using Neural Machine Translation (NMT) and Large Language Models (LLMs). The study found that post-editors frequently altered one in three metaphors, indicating significant issues with figurative language translation in literary contexts. Participants perceived the MT output quality as poor, often requiring more effort for post-editing than translating from scratch, which also impacted their creative freedom and sense of ownership. AI
IMPACT Highlights significant challenges in using LLMs and NMT for nuanced literary translation, suggesting current models may hinder rather than help creative post-editing.