Speeding up Transformer Decoding via an Attention Refinement Network
4767Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
Attention Operations
Method |
componentOf |
Transformer
Method |
|
Attention Refinement Network
Method |
enhances |
Transformer
Method |
|
Cross Attention
Method |
componentOf |
Transformer
Method |
|
Self Attention
Method |
componentOf |
Transformer
Method |
|
Weighted Residual Network
Method |
componentOf |
Attention Refinement Network
Method |
| evaluates |
Neural Machine Translation
Method |