search

Akeepers / LEAR

The implementation our EMNLP 2021 paper "Enhanced Language Representation with Label Knowledge for Span Extraction".
113 stars 13 forks source link

Visualization of the Proposed Model? #7

Closed yc1999 closed 2 years ago

yc1999 commented 2 years ago

Hi~in your paper, you show us a picture which indicates the non-effective attention mechanism of previous models:

image

I am wondering whether you can show us a picture like the above picture to demonstrate the effectiveness of LEAR's attention mechanism? Thank you~

Akeepers commented 2 years ago

Sorry, no such experiment has been done before.