Open WoshiBoluo opened 1 year ago
Hello, this paper of CrossTransformer is excellent, I have one question unable to solve. I found that the operation of cross-transformers and cross-attention are consistent, Is there any difference between them? Why is this module transformer-based?
Hello, this paper of CrossTransformer is excellent, I have one question unable to solve. I found that the operation of cross-transformers and cross-attention are consistent, Is there any difference between them? Why is this module transformer-based?