Metraipolis › archive › dc_3965d40fb252

Precision over Scale: A Polish-Silesian Benchmark and a Translation System Outperforming Open-Source and Commercial Models — arXiv:2610.01082

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.01082
License (as archived)
CC0-1.0
Archived
2026-10-06T13:54:51.478Z
SHA-256 of the archived text
02ef42a68c3a0ee6922ffd1afc5a2a16edd721ea9856e7ca7b6b032bee6ec394
Stable URL of this page
https://metraipolis.com/documento/dc_3965d40fb252

How to cite

Precision over Scale: A Polish-Silesian Benchmark and a Translation System Outperforming Open-Source and Commercial Models — arXiv:2610.01082 — arXiv (en), https://arxiv.org/abs/2610.01082 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3965d40fb252 (Metraipolis archive dc_3965d40fb252, sha256 02ef42a68c3a0ee6922ffd1afc5a2a16edd721ea9856e7ca7b6b032bee6ec394)
BibTeX
@misc{dc3965d40fb252,
  title = {Precision over Scale: A Polish-Silesian Benchmark and a Translation System Outperforming Open-Source and Commercial Models — arXiv:2610.01082},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01082},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3965d40fb252; sha256 02ef42a68c3a0ee6922ffd1afc5a2a16edd721ea9856e7ca7b6b032bee6ec394},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01082. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Dialectal machine translation remains challenging due to limited data and strong linguistic variation not captured by standard benchmarks, which often assume standardized and well-edited text. We study Polish-Silesian MT using neural and rule-based systems, evaluating on SiLTT - a new Pol-Szl testset, alongside established BOUQuET and FLORES benchmarks. Results show our rule-based system is consistently strongest on SiLTT and BOUQuET datasets and that TranslateGemma fine-tuned on a curated dataset improves over strong neural baselines but does not surpass the rule-based system in dialectal settings. We release SiLTT and our best neural model to support further research.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.01082 — 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 2026-10-06T12:15:34.676Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).