Hello,
We are happy to announce the WMT 2024 shared task on machine translation of Spanish texts
into the following underrepresented Romance languages of Spain: Aragonese, Aranese, and
Asturian.
In Spain, alongside Spanish, languages like Basque, Catalan, and Galician enrich the
linguistic landscape. While Spanish dominates machine translation resources, other
languages like Basque, Catalan, and Galician now provide substantial data for modern
systems. Yet, Spain also hosts truly low-resource languages like Aragonese, Aranese, and
Asturian, with far fewer resources, particularly in bilingual data.
The objectives of this shared task are to advance machine translation in resource-scarce
settings, investigate transferability among low-resource Romance languages when
translating from Spanish, optimize the utilization of pre-trained models for
Spanish-low-resource Romance language translation, and establish publicly accessible
corpora for machine translation development and evaluation.
Tailored datasets and language detectors have been released, along with the FLORES+
sentences, now available in these three languages and revised by official language
academies to align with current orthographic standards.
For more information please have a look at the shared task description:
https://www2.statmt.org/wmt24/romance-task.html
For questions please write to romance2...@dlsi.ua.es.
We are looking forwards to your submission!
Juan Antonio Pérez-Ortiz, Universitat d’Alacant
Felipe Sánchez-Martínez, Universitat d’Alacant
Antoni Oliver, Universitat Oberta de Catalunya
--
Universitat d'Alacant / Universidad de Alicante
Dept. de Llenguatges i Sistemes Informàtics
Felipe Sánchez Martínez
Associate Professor - Profesor Titular de Universidad
Tel.: (+34) 965 90 34 00 ext. 2966
Email: fsanc...@ua.es, fsanc...@dlsi.ua.es
Web: https://www.dlsi.ua.es/~fsanchez/
_______________________________________________
Apertium-stuff mailing list
Apertium-stuff@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/apertium-stuff