Sparse Distillation: Speeding Up Text Classification by Using Bigger Student Models
21238Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
Модели-ученики
Material |
evaluatedIn | |
|
Модели-ученики
Material |
includes |
Самовнимание
Method |
|
Модели-ученики
Material |
uses |
N-граммные встраивания
Material |