szhengac commented on pull request #19387: URL: https://github.com/apache/incubator-mxnet/pull/19387#issuecomment-713114122
> @MoisesHer Just FYI. The local attention will soon be the new standard of building transformers. Yes. We would need some help in optimizing the operator. Currently it saves significant memory for long sequence but not faster than full attention. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
