Translation Mt5 Small 128 En Hu

🚀 mT5 Translation model

A machine translation model that translates English to Hungarian, leveraging the mT5-small pre - trained model and fine - tuned on OPUS sub - corpora.

🚀 Quick Start

For further models, scripts and details, see our repository or our demo site.

✨ Features

• Language Pair: Translates from English to Hungarian.
• Pretrained Model: Utilizes the mT5 - small pre - trained model.
• Fine - Tuning: Fine - tuned on sub - corpora from OPUS with 56,837,602 segments.
• Prefix: Uses the prefix "translate English to Hungarian: ".

🔧 Technical Details

Limitations

• Input Requirement: Requires tokenized input text using the HuSpaCy tokenizer.
• Length Constraints: max_source_length = 128 and max_target_length = 128.

📚 Documentation

Results

📄 License

This project is licensed under the Apache - 2.0 license.

📚 Citation

If you use this model, please cite the following paper:

@inproceedings {laki-yang-mt,
    title = {{Jobban fordítunk magyarra, mint a Google!}},
    booktitle = {XVIII. Magyar Számítógépes Nyelvészeti Konferencia},
    year = {2022},
    publisher = {Szegedi Tudományegyetem, Informatikai Intézet},
    address = {Szeged, Magyarország},
    author = {Laki, László and Yang, Zijian Győző},
    pages = {357--372}
}