Trained on 100 million words and still in shape: BERT meets British National Corpus
33306Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
British National Corpus
Material |
comparesWith |
BERT
Method |
|
LTG BERT
Method |
improves |
BERT
Method |
|
MaskedLanguageModels
Method |
trainedOn |
British National Corpus
Material |