Keynote Speeches


Fake News 2.0: Combating Neural False Information

Abstract. The recent world-wide rise of “false information” (that may include rumor, clickbait, satire, fake news, hoax, misinformation, or disinformation) has caused significant confusion and disruption in societies. The recently-emerged phenomenon of “deepfakes” (AI-synthesized realistic artifacts) and other neural fakes could exacerbate the problem even further. In this talk, I will first present a review on the state-of-the-art computational solutions from my group as well as other leading groups that attempt to combat this false information and deepfakes. I will next discuss the important implications and impacts of “neural” false information and deepfakes in diverse applications, and conclude with final thoughts, emphasizing the needs of interdisciplinary and transdisciplinary approaches to this very challenging and important problem.

Bio. Dongwon Lee is an associate professor in the College of Information Sciences and Technology (aka iSchool) at Penn State University, and also an ACM distinguished member elected in 2019. Before starting at Penn State, he has worked at AT&T Bell Labs and obtained his Ph.D. in Computer Science from UCLA. From 2012 to 2015, he has served as a co-founder and chief scientist of Nittany Systems Research, and From 2015 to 2017, he has served as a program director at National Science Foundation, co-managing cybersecurity education programs with the yearly budget of $55M. In general, he researches the problems in the areas of data science, machine learning, and cybersecurity. Since 2017, in particular, he has led the SysFake project at Penn State, investigating to better understand and develop computational methods to combat fake news. More details of his research can be found at: http://pike.psu.edu/.

Revitalizing Local Journalism with AI-Assisted Writing: a Human-in-the-loop Approach

Abstract. In recent years, local journalism has been struggling to survive. The disappearance of thousands of local newspapers and laid off local journalists has left millions of people without a vital source of information, much of it being the critical original reporting. We at NewsBreak (an intelligent digital platform for local news), always believe in addressing the problem by empowering local journalists directly with human-in-the-loop AI technologies. In this talk, I will first present our study on how a local journalist would create a news piece, revealing several pain points. I will then discuss our internally developed assisted writing technologies designed to address such pain points along with their academic roots. Next, I will explain how local journalists would interact with our platform, while emphasizing the importance of human-in-the-loop to ensure the quality of local reporting. Last but not least, I would like to conclude by sharing several key open problems with the academic community which could help revitalizing local journalism.

Bio. Dr. Haoruo Peng currently serves as Head of Content Science at NewsBreak, the #1 intelligent news app & local information platform in US. He previously obtained his Ph.D. degree from University of Illinois at Urbana-Champaign in 2018, working with Prof. Dan Roth in the Cognitive Computation Group. Prior to that, he received his bachelor degree from Tsinghua University in 2013. His research interests lie mostly in the area of discourse understanding, such as co-reference resolution, event extraction and co-reference, discourse parsing and semantic language modeling (script learning). He has published research papers mainly in the NLP area, including ACL, EMNLP, and NAACL. He also serves as an area chair in ACL 2021, and has served as a PC member in ACL, EMNLP, NAACL, AAAI, ICML, IJCAI & NeurIPS.

Panel Discussion: The Future of News Intelligence

Huan Liu

Arizona State University

Dongwon Lee

Penn State University

Min Zhang

Tsinghua University

Haoruo Peng

NewsBreak

Accepted Papers

Does Gender Matter in the News? Detecting and Examining Gender Bias in News Articles

Jamell Dacon and Haochen Liu

Boosting Share Routing for Multi-task Learning

Xiaokai Chen, Xiaoguang Gu and Libo Fu

Collocating News Articles with Structured Web Tables

Alyssa Lees, Flip Korn, You Wu, Luciano Barbosa, Levy de Souza Silva and Cong Yu

Champion Solution in the MIND News Recommendation Challenge

Huige Cheng

How does Truth Evolve into Fake News? An Empirical Study of Fake News Evolution

Mingfei Guo, Xiuying Chen, Juntao Li, Dongyan Zhao and Rui Yan

Combining Explicit Entity Graph with Implicit Text Information for News Recommendation

Xuanyu Zhang and Qing Yang

Tweet Sentiment Analysis of the 2020 U.S. Presidential Election

Ethan Xia, Han Yue and Hongfu Liu

Toward the Next Generation of News Recommender Systems

Himan Abdollahpouri, Edward Malthouse, Joseph Konstan, Bamshad Mobasher and Jeremy Gilbert

EUDETECTOR: Leveraging Language Model to Identify EU-Related News

Koustav Rudra, Danny Tran and Miroslav Shaltev

Planned Schedule

Program Title UTC-4 UTC+8
Opening
Chair: Fangzhao Wu
Opening Remark 8:00-8:05 20:00-20:05
Keynote 1
Chair: Xing Xie
Fake News 2.0: Combating Neural False Information 8:05-8:40 20:05-20:40
Session 1
Chair: Xiang Ao
Long Paper Does Gender Matter in the News? Detecting and Examining Gender Bias in News Articles 8:40-8:55 20:40-20:55
Long Paper Collocating News Articles with Structured Web Tables 8:55-9:10 20:55-21:10
Short Paper How does Truth Evolve into Fake News? An Empirical Study of Fake News Evolution 9:10-9:20 21:10-21:20
Short Paper Combining Explicit Entity Graph with Implicit Text Information for News Recommendation 9:20-9:30 21:20-21:30
Short Paper Champion Solution in the MIND News Recommendation Challenge 9:30-9:40 21:30-21:40
Panel Discussion
Chair: Kai Shu
The Future of News Intelligence 9:40-10:25 21:40-22:25
Keynote 2
Chair: Ying Shan
Revitalizing Local Journalism with AI-Assisted Writing: a Human-in-the-loop Approach 10:25-11:00 22:25-23:00
Session 2
Chair: Chuhan Wu
Long Paper Boosting Share Routing for Multi-task Learning 11:00-11:15 23:00-23:15
Contributed Talk Random Walks with Erasure: Diversifying Personalized Recommendations on Social and Information Networks 11:15-11:30 23:15-23:30
Short Paper Tweet Sentiment Analysis of the 2020 U.S. Presidential Election 11:30-11:40 23:30-23:40
Short Paper Toward the Next Generation of News Recommender Systems 11:40-11:50 23:40-23:50
Short Paper EUDETECTOR: Leveraging Language Model to Identify EU-Related News 11:50-12:00 23:50-24:00
Closing Closing Remark 12:00-12:05 24:00-0:05