Title

Defending Against Neural Fake News

Summary

The paper discusses the dual-use concerns of natural language generation technology, highlighting how it can be exploited to create neural fake news that closely mimics real news. It introduces Grover, a model designed to both generate and detect such disinformation, demonstrating that the best defense against neural fake news might be strong public generators like Grover itself. The paper also explores the need for robust verification techniques and addresses ethical considerations in the public release of such technology.

 

 

 

 

َAuthor

Rowan Zellers, Ari Holtzman, Hannah Rashkin, Yonatan Bisk, Ali Farhadi, Franziska Roesner, Yejin Choi.

Year

2020

َThematic Area

Communication Studies

Topic

Fake information and Social media

Country

Global

Region

Global

Misinformation Combatting

Detection of Misinformation

Place Published

APA 7th End Text Citation

Rowan Zellers, Ari Holtzman, Hannah Rashkin, Yonatan Bisk, Ali Farhadi, Franziska Roesner, Yejin Choi. (2020). Defending Against Neural Fake News. https://doi.org/10.48550/arXiv.1905.12616