On the importance of pre-training data volume for compact language models

29825
Back to Articles Dashboard

Knowledge Triples

Source Entity Relationship Target Entity
trainedOn
comparesWith