Scaling Badini Kurdish POS Annotation with Hybrid Encoder–LLM Pre-Annotation and Agreement Routing
Journal of Applied Science and Technology Trends · 17.11.2026
DOI: 10.38094/jastt721293 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Applied Science and Technology Trends, online erschienen am 17.11.2026. Verfasst von Shahab Zebari und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
Badini Kurdish, a Kurmanji-group variety written in an unstandardized Arabic-based script, has a single manually POS-annotated resource: the UOZBDN corpus (920 sentences, 54,116 tokens, 38 tags).Expanding it toward the million-token corpus that Badini language technology requires is limited by expert annotation effort. This paper builds a hybrid pre-annotation pipeline and measures what it saves. On a frozen, sentence-disjoint test split we re-evaluate the corpus's rule-based + HMM tagger (82.81%), evaluate a large language model from zero-shot (78.92%) to retrieval-augmented prompting (82.38%, replicated at 82.43% by a second provider), and fine-tune the XLMRoBERTa encoder on the gold training split, which becomes the best single system (85.41%). Because the encoder and the retrievalgrounded LLM err on different tokens, the pipeline auto-accepts tokens on which they agree (83.10% of tokens, 92.44% accurate) and routes disagreements to a human reviewer. A control experiment shows that agreement selects no more accurately than the encoder's own confidence at matched coverage (92.49% vs. 92.44%); what it buys is a denser review interface, worth about a single model's top-2.65 candidates. At an illustrative reviewer accuracy, the routed corpus reaches a projected 90.60% under two-candidate review and 93.29% under free entry, with 16.90% of tokens routed for review (a token share, not a measured effort saving) and $159–$317 of API cost per million tokens. A leave-one-genre-out study on an encoder–HMM pairing shows agreed-token accuracy falling 91.34% to 81.17%, so the projection holds in-domain only. Code, model outputs, and the corpus are available on request.
Zitieren
APA
Zebari, S., & Jacksi, K. (2026). Scaling Badini Kurdish POS Annotation with Hybrid Encoder–LLM Pre-Annotation and Agreement Routing. Journal of Applied Science and Technology Trends. https://doi.org/10.38094/jastt721293
Vancouver
Zebari S, Jacksi K. Scaling Badini Kurdish POS Annotation with Hybrid Encoder–LLM Pre-Annotation and Agreement Routing. Journal of Applied Science and Technology Trends. 2026. doi:10.38094/jastt721293
BibTeX
@article{Zebari2026,
author = {Zebari, Shahab and Jacksi, Karwan},
title = {Scaling Badini Kurdish POS Annotation with Hybrid Encoder–LLM Pre-Annotation and Agreement Routing},
journal = {Journal of Applied Science and Technology Trends},
year = {2026},
doi = {10.38094/jastt721293}
}
EUVIMED