Fluency and Faithfulness in Human and Machine Literary Translation
Sarah Griebel, Ted Underwood, [email protected] · 2026
Read Fluency and Faithfulness in Human and Machine Literary Translation by Sarah Griebel, Ted Underwood, [email protected] (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-15282.txt.
Description
Sarah Griebel and Ted Underwood examine the tradeoff between fluency and loyalty in literary translation across 130,486 paragraphs from 106 novels in 16 tongues. Fluency is gauged as source-closeness via a translationese sorter, loyalty via the COMET-KIWI metric, holding for passage length. A steady inverse link emerges: more fluent passages tend to be less faithful. The pattern holds for human and Google Translate output but weakens for TranslateGemma, and segment length matters for auto evaluation.
Catalog record
| CREATOR | Sarah Griebel, Ted Underwood, [email protected] |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33204 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.15282