# Multilingual Fine-tuning

An Icelandic masked language model based on the RoBERTa-base architecture, fine-tuned from xlm-roberta-base

This model is a text summarization model fine-tuned on the Amazon dataset based on google/mt5-small, supporting English to Traditional Chinese abstract generation tasks.

A cross-lingual sentence embedding model supporting English and German, applicable for tasks such as semantic textual similarity, semantic search, and paraphrase mining.

A German automatic speech recognition model fine-tuned on the MULTILINGUAL_LIBRISPEECH - GERMAN dataset, based on facebook/wav2vec2-xls-r-100m

English-Romanian translation model based on mbart-large-cc25 architecture, fine-tuned on the wmt_en_ro dataset

A Hausa language model fine-tuned based on XLM-RoBERTa, outperforming the base model in text classification and named entity recognition tasks

This is a Hausa BERT model fine-tuned based on the multilingual BERT model, performing better than the original multilingual BERT in text classification and named entity recognition tasks.