Metraipolis › archive › dc_440eefd2c389

Augmenting Text to Increase Translation Difficulty — arXiv:2608.15932

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.15932
License (as archived)
CC0-1.0
Archived
2026-10-18T17:33:26.005Z
SHA-256 of the archived text
391fe9706e8d0d18a203805a816c3d98954baa4b8a276ae0f54de14969ce885c
Stable URL of this page
https://metraipolis.com/documento/dc_440eefd2c389

How to cite

Augmenting Text to Increase Translation Difficulty — arXiv:2608.15932 — arXiv (en), https://arxiv.org/abs/2608.15932 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_440eefd2c389 (Metraipolis archive dc_440eefd2c389, sha256 391fe9706e8d0d18a203805a816c3d98954baa4b8a276ae0f54de14969ce885c)
BibTeX
@misc{dc440eefd2c389,
  title = {Augmenting Text to Increase Translation Difficulty — arXiv:2608.15932},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15932},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_440eefd2c389; sha256 391fe9706e8d0d18a203805a816c3d98954baa4b8a276ae0f54de14969ce885c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15932. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As state-of-the-art machine translation models saturate standard benchmarks, the field needs more challenging evaluations to distinguish between models of varying quality. We propose augmenting existing benchmarks to increase translation difficulty by combining adversarial optimization with a differentiable translation difficulty estimator. Our Adversarial Translation Optimization (ATO) uses gradients from a combined difficulty and fluency objective to iteratively replace tokens. Because each step branches over candidate substitutions at every position, optimization becomes a tree search problem, which we address with Beam Search. ATO offers a gradient-based alternative to LLM-based dataset creation without LLM prompting, expensive human curation, or task-specific model training. Our ATO-modified benchmark lowers average translation quality (xCOMET) from 0.93 to 0.82, compared to 0.88 for paraphrasing and 0.86 for a zero-shot baseline. Human evaluation shows the modified texts are somewhat less natural than the baselines but remain reasonably grammatical and plausible while being substantially harder to translate. We release two datasets of 350 English texts each, generated by our methods, as well as the code.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.15932 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T16:29:13.842Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).