Login / Signup

OFA: A Framework of Initializing Unseen Subword Embeddings for Efficient Large-scale Multilingual Continued Pretraining.

Yihong LiuPeiqin LinMingyang WangHinrich Schütze
Published in: CoRR (2023)
Keyphrases
  • lightweight
  • theoretical framework
  • database
  • data sets
  • genetic algorithm
  • main contribution
  • real world
  • multimedia
  • semi supervised
  • low dimensional
  • euclidean space