Machine translation meets large language models : evaluating ChatGPT’s ability to automatically post-edit literary texts

Publication type
C1
Publication status
Published
Author
Macken, L.
Editor
Bram Vanroy, Marie-Aude Lefer, Lieve Macken and Paola Ruffo
Series
Proceedings of the First Workshop on Creative-text Translation and Technology
Pagination
71-87
Conference
1st workshop on Creative-text Translation and Technology (CTT), co-located with the 25th Annual Conference of the European Association for Machine Translation (EAMT 2024) (Sheffield, UK)
Download
(.pdf)
View in Biblio
(externe link)

Abstract

Large language models such as GPT-4 have been trained on vast corpora, giving them excellent language understanding. This study explores the use of Chat-GPT for post-editing machine translations of literary texts. Three short stories, machine translated from English into Dutch, were post-edited by 7-8 professional translators and ChatGPT. Automatic metrics were used to evaluate the number and type of edits made, and semantic and syntactic similarity between the machine translation and the corresponding post-edited versions. A manual analysis classified errors in the machine translation and changes made by the post-editors. The results show that ChatGPT made more changes than the average post-editor. ChatGPT improved lexical richness over machine translation for all texts. The analysis of editing types showed that ChatGPT replaced more words with synonyms, corrected fewer machine errors and introduced more problems than professionals.