BlAInded by Fluency: How Idiomatic Machine Translation Outputs Affect Student Post-Editors’ Edit Types

Valentin Scourneau, Loïc De Faria Pires


Abstract
This study explores the influence of two prompting strategies on the lexical and syntactic metrics of the LLM-powered machine translations (MTs) of a corpus of 18~British editorials into French as well as their impact on the edit types made by Master’s translation students post-editing from a representative editorial of the corpus, as evaluated using the MTPEAS taxonomy. Quantitatively, the prompt specifically requesting more syntactic and lexical variety leads to significantly higher syntactic and lexical metrics scores in the MTs, but differences remain significant only for lexical metrics in the post-edited versions of the representative editorial. Qualitatively, we show that students post-editing from an MT featuring more idiomatic rephrasings and fewer syntactic calques (as opposed to an MT that is structurally closer to the source text) seem to make fewer edits overall, leave more MT errors unaddressed, and make fewer successful edits.
Anthology ID:
2026.eamt-1.56
Volume:
Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 1)
Month:
June
Year:
2026
Address:
Tilburg, The Netherlands
Editors:
Dimitar Shterionov, Eva Vanmassenhove, Mirella De Sisto, Fred Blain, Javad Pourmostafa Roshan Sharami, Lisa Lepp, Chiara Manna, Argentina Anna Rescigno, Alina Karakanta, Ayla Rigouts Terryn, Manuel Lardelli, Natalia Resende, Elena Murgolo, Janiça Hackenbuchner, Anna Zaretskaya, Miquel Esplà-Gomis, Thierry Etchegoyhen, Dagmar Gromann, Rachel Bawden, Barry Haddow, Sara Szoc, Mikel Forcada, Helena Moniz
Venue:
EAMT
SIG:
Publisher:
European Association for Machine Translation
Note:
Pages:
869–882
Language:
URL:
https://aclanthology.org/2026.eamt-1.56/
DOI:
Bibkey:
Cite (ACL):
Valentin Scourneau and Loïc De Faria Pires. 2026. BlAInded by Fluency: How Idiomatic Machine Translation Outputs Affect Student Post-Editors’ Edit Types. In Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 1), pages 869–882, Tilburg, The Netherlands. European Association for Machine Translation.
Cite (Informal):
BlAInded by Fluency: How Idiomatic Machine Translation Outputs Affect Student Post-Editors’ Edit Types (Scourneau & Pires, EAMT 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.eamt-1.56.pdf