HILANCO

Hungarian Intelligent Language Applications Consortium

« Back to Home Page

HILBART

  • HILBART large web: Trained on 1% of Webcorpus 2.0.
  • HILBART base web: Trained on 10% of Webcorpus 2.0.
  • HILBART base wiki: Trained on Hungarian Wikipedia.

References

More Language Models

HIL-ELECTRA
» HIL-ELECTRA NYTI
- trained on NYTI-BERT corpus
» HIL-ELECTRA wiki
- trained on Hungarian Wikipedia

HIL-RoBERTa
» HIL-RoBERTa wiki
- trained on Hungarian Wikipedia

HIL-ALBERT
» HIL-ALBERT NYTI
- trained on 10% of NYTI-BERT corpus
» HIL-ALBERT wiki
- trained on Hungarian Wikipedia

HIL-SBERT
Sentence-BERT models are fine-tuned BERT networks aimed at obtaining high-quality sentence embeddings.