Metraipolis › archive › dc_cbddf82f7186

The University of Melbourne WMT 2026 CreoleMT Submission: A Domain-Balanced Approach to Low-Resource Pacific Creole Machine Translation — arXiv:2609.13615

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13615
License (as archived)
CC0-1.0
Archived
2026-10-15T23:13:28.736Z
SHA-256 of the archived text
99ae5ca9f2a3997291782775e3a0ed5065a4ea2605e7340ab4ba0ae161c65541
Stable URL of this page
https://metraipolis.com/documento/dc_cbddf82f7186

How to cite

The University of Melbourne WMT 2026 CreoleMT Submission: A Domain-Balanced Approach to Low-Resource Pacific Creole Machine Translation — arXiv:2609.13615 — arXiv (en), https://arxiv.org/abs/2609.13615 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cbddf82f7186 (Metraipolis archive dc_cbddf82f7186, sha256 99ae5ca9f2a3997291782775e3a0ed5065a4ea2605e7340ab4ba0ae161c65541)
BibTeX
@misc{dccbddf82f7186,
  title = {The University of Melbourne WMT 2026 CreoleMT Submission: A Domain-Balanced Approach to Low-Resource Pacific Creole Machine Translation — arXiv:2609.13615},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13615},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cbddf82f7186; sha256 99ae5ca9f2a3997291782775e3a0ed5065a4ea2605e7340ab4ba0ae161c65541},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13615. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

For our submission to the WMT26 Creole Language Translation Shared Task, we focus on machine translation (MT) models for Pacific creoles: Tok Pisin, Bislama, and Solomon Pijin, with particular attention to broad domain performance. After pre-training on a large collection of domain-imbalanced data, we continue fine-tuning on a diverse mix of domain-balanced data. We rely on a number of data collection and preparation techniques, including LLM-assisted respelling and alignment, back-translation, and distillation from Gemini for domains originally not present in training data. Evaluated on Bouquet and a novel test set made of spoken language transcripts, our models beat open model baselines by 3+ chrF++ points in all directions with human-original references. Looking ahead, we plan to develop human-translated test sets for Solomon Pijin and Bislama, and to distil our best models into much smaller ones that retain broad domain coverage.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.13615 — 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 2026-10-15T22:51:09.829Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).