Adversary-Aware Rumor Detection

While social media becomes a primary source of news now, it also becomes more challenging for people to distinguish the rumors and non-rumors, which attracts malicious manipulation and may lead to public health harm or economic loss. Consequently, many rumor detection models have been proposed to automatically detect the rumors based on the contents and propagation path. However, most previous works are not aware of malicious attacks, e.g., framing. Therefore, we propose a novel rumor detection framework, Adversary-Aware Rumor Detection including Weighted-Edge Transformer-Graph Network and Position-aware Adversarial Response Generator, to improve the vulnerability of detection models. To the best of our knowledge, this is the first work that can generate the adversarial response with the consideration of the response position. Experimental results show that our model achieves the state-of-theart on various rumor detection tasks by the proposed Weighted-Edge Transformer-Graph Network and can maintain the performance under the adversarial response attack after the adversarial learning by Position-aware Adversarial Response Generator.1

[1]  Katherine Del Giudice.  Crowdsourcing credibility: The impact of audience feedback on Web page credibility , 2010, ASIST.

[2]  Wei Gao,et al.  Rumor Detection on Twitter with Tree-structured Recursive Neural Networks , 2018, ACL.

[3]  Ben Poole,et al.  Categorical Reparameterization with Gumbel-Softmax , 2016, ICLR.

[4]  Cheng-Te Li,et al.  GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media , 2020, ACL.

[5]  Xiaoyu Yang,et al.  Rumor Detection on Social Media with Graph Structured Adversarial Learning , 2020, IJCAI.

[6]  LiakataMaria,et al.  Detection and Resolution of Rumours in Social Media , 2018 .

[7]  Ashish Vaswani,et al.  Self-Attention with Relative Position Representations , 2018, NAACL.

[8]  Wei Gao,et al.  Detecting Rumors from Microblogs with Recurrent Neural Networks , 2016, IJCAI.

[9]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[10]  Wenji Mao,et al.  Modeling Conversation Structure and Temporal Dynamics for Jointly Predicting Rumor Stance and Veracity , 2019, EMNLP.

[11]  Wei Gao,et al.  Detect Rumors in Microblog Posts Using Propagation Structure via Kernel Learning , 2017, ACL.

[12]  Jing Jiang,et al.  Interpretable Rumor Detection in Microblogs by Attending to User Interactions , 2020, AAAI.

[13]  Wenbing Huang,et al.  Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks , 2020, AAAI.

[14]  Mirella Lapata,et al.  Text Summarization with Pretrained Encoders , 2019, EMNLP.

[15]  Wei Gao,et al.  Detect Rumors on Twitter by Promoting Information Campaigns with Generative Adversarial Learning , 2019, WWW.

[16]  Aditi Raghunathan,et al.  Adversarial Training Can Hurt Generalization , 2019, ArXiv.

[17]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[18]  Shakir Mohamed,et al.  Training language GANs from Scratch , 2019, NeurIPS.

[19]  Benno Stein,et al.  A Stylometric Inquiry into Hyperpartisan and Fake News , 2017, ACL.

[20]  Arkaitz Zubiaga,et al.  Learning Reporting Dynamics during Breaking News for Rumour Detection in Social Media , 2016, ArXiv.

[21]  Xiaomo Liu,et al.  Real-time Rumor Debunking on Twitter , 2015, CIKM.

[22]  Jie Gao,et al.  Data Augmentation for Rumor Detection Using Context-Sensitive Neural Language Model With Large-Scale Credibility Corpus , 2019 .

[23]  Svitlana Volkova,et al.  Separating Facts from Fiction: Linguistic Models to Classify Suspicious and Trusted News Posts on Twitter , 2017, ACL.

[24]  Sameer Badaskar,et al.  Identifying Real or Fake Articles: Towards better Language Modeling , 2008, IJCNLP.

[25]  Luo Si,et al.  Rumor Detection by Exploiting User Credibility Information, Attention and Multi-task Learning , 2019, ACL.

[26]  Barbara Poblete,et al.  Information credibility on twitter , 2011, WWW.

[27]  Kathleen M. Carley,et al.  Tree LSTMs with Convolution Units to Predict Stance and Rumor Veracity in Social Media Conversations , 2019, ACL.

[28]  P. Bordia,et al.  Rumor, Gossip and Urban Legends , 2007 .