Enhancing Task-Specific Distillation in Small Data Regimes through Language Generation

4701

Knowledge Triples

Source Entity	Relationship	Target Entity
BERT Method	usedFor	Data Augmentation Method
DBpedia Material	evaluatedIn	Data Augmentation Method
KnowledgeDistillation Method	addresses	Large Scale Pre Trained Language Models Material
KnowledgeDistillation Method	uses	PseudoLabels Material
Large Scale Pre Trained Language Models Material	improves	Natural Language Processing OtherEntity
Monte Carlo Tree Search Method	appliesTo	Synthetic Data Generation Method
MRPC Material	evaluatedIn	Data Augmentation Method
Natural Language Generation Task	involves	Synthetic Data Generation Method
Non Targeted Text Generation Method	appliesTo	Synthetic Data Generation Method
SST 2 Material	evaluatedIn	Data Augmentation Method
Synthetic Data Generation Method	enhances	KnowledgeDistillation Method
TREC 6 Material	evaluatedIn	Data Augmentation Method
YELP 2 Material	evaluatedIn	Data Augmentation Method