Is there any plan to share paper or tutorial ?

Tarpelite commented 4 years ago

Hi, I'm really interested in your work and I have recently worked at using BERT to do parsing, But I am a little fresh about how to deal with the tokenizing and the decoding,

So do you have any plan to release a paper or do any tutorial?

LiangsLi commented 4 years ago

Hi， @Tarpelite I am very honored to know that you are interested in this project. Since this project has a lot of follow-up work to do, and I have other work in progress, we have no plans to write a tutorial in the near future. Although we haven't published a paper yet, there is a lot of similar work at EMNLP 2019 and other conferences. You can refer to the following papers:

Beto, Bentz, Becas: The Surprising Cross-Lingual Effectiveness of BERT
Multilingual Constituency Parsing with Self-Attention and Pre-Training
75 Languages, 1 Model: Parsing Universal Dependencies Universally
Cross-Lingual BERT Transformation for Zero-Shot Dependency Parsing

In addition, Stanford has released a great dependency parser, and our decoder also refers to their implementation, you can refer to stanfordnlp.

hope this helps~

Tarpelite commented 4 years ago

Thanks for your very detailed helpful suggestions. 😃

NLPInBLCU / BiaffineDependencyParsing

Is there any plan to share paper or tutorial ? #7