Gwo-K Kreyol Classifier: Rock Pidgins (RP) Submission to the WMT 2026 CreoleMT Language Identification Shared Task

Emmett Strickland


Abstract
This paper describes ongoing efforts to build an n-gram text classifier for several major French-lexifier creoles. Our model is based on a simple text processing pipeline, limited training data, and a deliberately lightweight architecture. Despite this, it achieves substantially stronger performance than existing baselines on our target languages. These results suggest that long-established machine learning approaches remain effective for new language identification needs.
Anthology ID:
2026.wmt-main.191
Volume:
Proceedings of the Eleventh Conference on Machine Translation
Month:
October
Year:
2026
Address:
Budapest, Hungary
Editors:
Barry Haddow, Tom Kocmi, Philipp Koehn, Christof Monz
Venues:
WMT | WS
SIG:
SIGMT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2471–2475
Language:
URL:
https://aclanthology.org/2026.wmt-main.191/
DOI:
Bibkey:
Cite (ACL):
Emmett Strickland. 2026. Gwo-K Kreyol Classifier: Rock Pidgins (RP) Submission to the WMT 2026 CreoleMT Language Identification Shared Task. In Proceedings of the Eleventh Conference on Machine Translation, pages 2471–2475, Budapest, Hungary. Association for Computational Linguistics.
Cite (Informal):
Gwo-K Kreyol Classifier: Rock Pidgins (RP) Submission to the WMT 2026 CreoleMT Language Identification Shared Task (Strickland, WMT 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.wmt-main.191.pdf