# Multilingual BERT Fine-tuning

Model fine-tuned based on bert-base-multilingual-cased, achieving 99.93% accuracy on the evaluation set

Marathi named entity recognition model fine-tuned from bert-base-multilingual-cased

This is a pre-trained model for detecting unassimilated English lexical borrowings (also known as Anglicisms) in Spanish news.

A bilingual (English + German) QA model fine-tuned on the bert-base-multilingual-cased model, supporting SQuAD2.0 and deQuAD2.0 datasets

This is a BERT model optimized for Nigerian Pidgin, fine-tuned from the multilingual BERT, and it outperforms the original multilingual BERT in named entity recognition tasks.