CINO: Pre-trained Language Models for Chinese Minority Languages(中国少数民族预训练模型)

Multilingual Pre-trained Language Model, such as mBERT, XLM-R, provide multilingual and cross-lingual ability for language understanding. We have seen rapid progress on building multilingual PLMs in recent year. However, there is a lack of contributions on building PLMs on Chines minority languages, which hinders researchers from building powerful NLP systems.

To address the absence of Chinese minority PLMs, Joint Laboratory of HIT and iFLYTEK Research (HFL) proposes CINO (Chinese-miNOrity pre-trained language model), which is built on XLM-R with additional pre-training using Chinese minority corpus, such as

  • Chinese,中文(zh)
  • Tibetan,藏语(bo)
  • Mongolian (Uighur form),蒙语(mn)
  • Uyghur,维吾尔语(ug)
  • Kazakh (Arabic form),哈萨克语(kk)
  • Korean,朝鲜语(ko)
  • Zhuang,壮语
  • Cantonese,粤语(yue)

Please read our GitHub repository for more details (Chinese):

You may also interested in,

Chinese MacBERT:
Chinese BERT series:
Chinese ELECTRA:
Chinese XLNet:
Knowledge Distillation Toolkit - TextBrewer:

More resources by HFL:

Downloads last month
Hosted inference API
Mask token: <mask>
This model can be loaded on the Inference API on-demand.