Tree-Transformer icon indicating copy to clipboard operation
Tree-Transformer copied to clipboard

Questions in paper

Open wang304381190 opened this issue 3 years ago • 0 comments

I found that the value of attention to oneself in the attention figure is 0, that is, the value of the diagonal part is 0, which means that one cannot attend to oneself. Will this affect performance? Is there any experimental verification? Thanks.

wang304381190 avatar Mar 31 '21 12:03 wang304381190