English

KyrgyzBERT: A Compact, Efficient Language Model for Kyrgyz NLP

Computation and Language 2025-11-26 v1

Abstract

Kyrgyz remains a low-resource language with limited foundational NLP tools. To address this gap, we introduce KyrgyzBERT, the first publicly available monolingual BERT-based language model for Kyrgyz. The model has 35.9M parameters and uses a custom tokenizer designed for the language's morphological structure. To evaluate performance, we create kyrgyz-sst2, a sentiment analysis benchmark built by translating the Stanford Sentiment Treebank and manually annotating the full test set. KyrgyzBERT fine-tuned on this dataset achieves an F1-score of 0.8280, competitive with a fine-tuned mBERT model five times larger. All models, data, and code are released to support future research in Kyrgyz NLP.

Keywords

Cite

@article{arxiv.2511.20182,
  title  = {KyrgyzBERT: A Compact, Efficient Language Model for Kyrgyz NLP},
  author = {Adilet Metinov and Gulida M. Kudakeeva and Gulnara D. Kabaeva},
  journal= {arXiv preprint arXiv:2511.20182},
  year   = {2025}
}

Comments

3 pages, 1 figure, 2 tables. Preprint

R2 v1 2026-07-01T07:54:01.792Z