BART Translation model

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

• Source language: English

• Target language: Hungarian

• BART base model:

  • Pretrained on English WikiText-103 and Hungarian Wikipedia
  • Finetuned on subcorpora from OPUS
    • Segments: 56.837.602

Limitations

• tokenized input text (tokenizer: HuSpaCy)
• max_source_length = 128
• max_target_length = 128

Results

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}
}