Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

这篇论文的思路非常新颖,且具备独创性,但结果似乎差强人意 #2

Closed
lexmen318 opened this issue Nov 28, 2019 · 2 comments

Comments

@lexmen318
Copy link

从论文的实现思路来看,应该还是比较清晰的。

但从实验结果来看,似乎在特定领域上相比BERT的性能提升幅度不明显,不知道是否深入分析过原因?

感谢!

@autoliuweijie
Copy link
Owner

确实,我们发现K-BERT在很多任务上效果不显著,我们也在分析其中的原因,可能原因有:

  1. 目前用于测评的任务并非是“知识驱动”的任务,这些数据在标注时并不需要背景知识,引入知识反而被认为标注“错误”;
  2. 知识图谱的质量不高,目前我们的知识图谱大多从开发领域图谱中筛选的,其中的知识BERT通过大规模语料预训练也能得到。

不得不承认,K-BERT还是存在很多问题,我们还在进一步优化和改进,谢谢您的关注。

@lexmen318
Copy link
Author

Github很奇怪,这问题记得之前已经关闭了。感觉好像穿越了!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants