fastHan icon indicating copy to clipboard operation
fastHan copied to clipboard

Chinese Dependency Parsing的SOTA模型

Open hl0737 opened this issue 4 years ago • 1 comments

作者您好:

我看了一下chinese dependency parsing在paperswithcode上的SOTA模型

显示是这一篇Glyce: Glyph-vectors for Chinese Character Representations

它的dependency parsing的分数89 L , 90.2 U, 而本项目readme中的文档中写的dependency parsing的SOTA的准确率没有这个高, 请问readme中是参考的哪篇的数据?

hl0737 avatar Dec 14 '21 10:12 hl0737

我们和Glyce用的数据集不一样,我们参考的是这篇: A graph-based model for joint chinese word segmentation and dependency parsing

fdugzc avatar Dec 15 '21 06:12 fdugzc