The Nawatl corpus PI-YALLI contains a set of several Nawatl documents, and a three different static embeddings models. The corpus was splitted in 16 topics.
The corpus was manually compiled and processed by a team from Université d'Avignon (France), Universidad Veracruzana (Mexique) and independent researchers.
This corpus is suitable for testing and learning systems working on Nawatl language. New versions, containing more texts, will be aggregated periodically.
The PI-YALLI corpus and its embeddings is distributed by Laboratoire Informatique d'Avignon (France) under LGPL license.
- Updated Pi-yalli Corpus Version 1.8 (soon available!)
- Télécharger/Bajar/Download le corpus nahuatl PI-YALLI / the Nahuatl corpus PI-YALLI / Nahuazösisch PI-YALLI Korpus
- Updated Embeddings pi-yalli Version 0.001 of 01.07.2025
How to cite this corpus ?
- Guzman-Landa, et al. (2025) PI-yalli: un nouveau corpus pour le nahuatl / Yankuik nawatlahtolkorpus pampa tlahtolmachiotl. [article] [bib] [résumé] TALN'25, Marseille, 2025
- Torres-Moreno JM., et al. (2024) NAHU2: Un nouveau corpus pour le Nahuatl , HAL 2024
- Torres-Moreno JM., et al. (2024) PI-yalli: un nouveau corpus pour le nahuatl. ArXiv, arXiv:2412.15821, https://doi.org/10.48550/arXiv.2412.15821
Contact : Juan-Manuel Torres-Morenohttp://lia.univ-avignon.fr / Universite d'Avignon, France
juan *-* manuel *dot* torres *at* univ-avignon *dot* fr
Updated 20.06.25