EMNLP-2019/11-Different Absorption from the Same Sharing: Sifted Multi-task Learning for Fake News Detection

[BrambleXu]

Summary:

在transformer之后添加attention和gate，来给不同task学习不同的feature。两个task分别是 Stance Detection 和 Fake News Detection。

Resource:

• pdf
• [code](
• [paper-with-code](

Paper information:

• Author:
• Dataset: RumourEval and PHEME,
• keywords:

Notes:

Fake news可能造成非常大的危害，How Does One Fake Tweet Cause a Stock Market Crash?

因为fake news经常会引发对于real news的讨论，所以 stance analysis of these controversies can serve as valuable credibility features for fake news detection. 所以把这两个task合起来了。

Model Graph:

Result:：

Thoughts:

Next Reading: