# AI translation of literary texts is "fine", but readers still prefer human translations

arXiv: https://arxiv.org/abs/2606.26040 · HTML: https://mlsanity.com/paper/2606.26040/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 60 references that are in the map.
- First posted on arXiv 24 Jun 2026.

## Builds on (most cited first, 4 of 4)

- [Technical Report on the Pangram AI-Generated Text Classifier](https://mlsanity.com/paper/2402.14873/) (2024)
- [A 2-step Framework for Automated Literary Translation Evaluation: Its Promises and Pitfalls](https://mlsanity.com/paper/2412.01340/) (2024)
- [Creativity Bias: How Machine Evaluation Struggles with Creativity in Literary Translations](https://mlsanity.com/paper/2605.13596/) (2026)
- [Loong: A Human-Like Long Document Translation Agent with Observe-and-Act Adaptive Context Selection](https://mlsanity.com/paper/2605.30274/) (2026)

## Built on by (most cited first, 1 of 1)

- [Contrastive ESA: Human Evaluation of Multiple Translations at Once](https://mlsanity.com/paper/2607.26640/) (2026)

## Abstract (arXiv)

AI translation of literary works is increasingly common. While the content may be rendered adequately, we do not know enough about how readers experience it in terms of immersiveness and literary effect-aspects poorly captured by automatic metrics or human evaluation targeting fluency and adequacy. We ask 15 avid readers to compare recently published human translations (HT) to machine translations (MT) generated with an agentic language model-based pipeline, for 15 recent novels in French, Polish, and Japanese translated into English. Readers evaluated approximately 8K-word excerpts in two conditions: immersive reading of the whole excerpt (30 comparisons) and close reading of 386 aligned HT-MT chunk pairs (772 comparisons), with two readers per book. Overall, readers find MT "fine", but prefer HT (descriptively at the excerpt level, 19/30, and significantly at the chunk level, 522/772) for its ease, clarity, and immersive nature. Readers' highlights show that MT's quality varies more within one book than HT's does. Crucially, readers cannot reliably tell the two apart (17/30 guess correctly) and tend to prefer the version they believe to be human. Automatic metrics, including LLM-as-a-judge approaches, fail to recover reader preferences and favor MT. We release LAIT (Literary AI Translation), a reader-centered evaluation dataset with 1K reader comments, 2K judgments and preference ratings, and 7.2K span-level annotations, along with our evaluation protocol and supporting interface.