NLP-progress
NLP-progress copied to clipboard
Add new results and a new dataset for SRL
Add new results and a new dataset for SRL.
@yuanheTian To be fair there are better SOTA nowadays https://paperswithcode.com/sota/semantic-role-labeling-on-ontonotes https://paperswithcode.com/sota/semantic-role-labeling-on-conll-2005 but thank you for the PR, also you are right to use XLnet, it is the best baseline, and I strongly believe that if XLnet was combined with the method in https://paperswithcode.com/paper/semantic-role-labeling-as-dependency-parsing it would give us a new #1 SOTA. (CRF2o) BTW it's related to your work on dependency parsing.
Note: I don't have the rights to merge PRs.
Thanks for adding this PR!