SaTC: Core: Small: Targeting Challenges in Computational Disinformation Research to Enhance Attribution, Detection, and Explanation

Description

The use of social media has accelerated information sharing and instantaneous communications. The low barrier to enter social media enables more users to participate and makes them stay engaged longer, incentivizing individuals with a hidden agenda to spread disinformation online to manipulate information and sway opinion. Disinformation, such as fake news, hoaxes, and conspiracy theories, has increasingly become a hindrance to the functioning of online social media as an effective channel for trustworthy information. Cases are emerging where deliberately-fabricated disinformation is weaponized to divide people and create detrimental societal effects. Therefore, it is imperative to understand disinformation and to systematically investigate how we can improve resistance against it, taking into account the tension between the need for information and the need for security and protection from disinformation. The goal of the proposed research is to study the scientific underpinnings of disinformation and develop a computational framework to attribute, detect, and explain disinformation to inform policymaking. Developing such a framework depends greatly on close interaction across interdisciplinary theories, machine learning, and social media mining, which this project intends to unify. The project will develop fundamental improvements and develop new knowledge and a systematic computational framework to address major (provenance, data, and explainability) challenges when tackling online disinformation.

Publications

• Canyu Chen, Kai Shu. ``Combating Misinformation in the Age of LLMs: Opportunities and Challenges'', AI Magazine, AAAI, 2024.
• Kay Liu, Yingtong Dou, Yue Zhao, Xueying Ding, Xiyang Hu, Ruitong Zhang, Kaize Ding, Canyu Chen, Hao Peng, Kai Shu, Lichao Sun, Jundong Li, George H. Chen, Zhihao Jia, Philip S. Yu. ``PyGOD: A Python Library for Graph Outlier Detection", Journal of Machine Learning Research (JMLR), 2023.

• Xiongxiao Xu, Kaize Ding, Canyu Chen and Kai Shu. ``MetaGAD: Meta Representation Adaptation for Few-Shot Graph Anomaly Detection'', Proceedings of The 11th IEEE International Conference on Data Science and Advanced Analytics (DSAA 2024).
• Haoran Wang, Kai Shu. ``Backdoor Activation Attack: Attack Large Language Models using Activation Steering for Safety-Alignment'', Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (CIKM 2024).
• Sj Dillon*, Yueqing Liang*, H. Russell Bernard, Kai Shu. ``Investigating Gender Euphoria and Dysphoria on TikTok: Characterization and Comparison,'', Proceedings of 16th International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2024).
• Junwei Yin, Min Gao, Kai Shu, Jia Wang, Yinqiu Huang, Wei Zhou. ``Fine-Grained Discrepancy Contrastive Learning for Robust Fake News Detection", Proceedings of 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2024).
• Haoran Wang, and Kai Shu.``Explainable Claim Verification via Knowledge-Grounded Reasoning with Large Language Models", Findings of The 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023-Findings).
• Hao Liao, Jiahao Peng, Zhanyi Huang, Wei Zhang, Guanghua Li, Kai Shu, and Xing Xie. ``MUSER : A MUlti-Step Evidence Retrieval Enhancement Framework for Fake News Detection", Proceedings of 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023), ADS Track.
• Haoran Wang, Yingtong Dou, Canyu Chen, Lichao Sun, Philip S. Yu and Kai Shu. ``Attacking Fake News Detectors via Manipulating News Social Engagement", Proceedings of The 2023 ACM Web Conference (WWW 2023).
• Canyu Chen, and Kai Shu. ``PromptDA: Label-guided Data Augmentation for Prompt-based Few Shot Learners", Proceedings of The 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2023).

Resources

• [AttackFakeNews]: Attacking fake news detection via manipulating social engagements
• [PromptDA]: Prompt-based few-shot text classification
• [FOLK]: Leveraging LLMs to enhance fact-checking
• [MUSER]: A multi-step evidence retrieval enhancement framework for fake news detection
• [PyGOD]: An open-source toolkit for graph anomaly detection in Python

• [FakeNewsNet]: Fake news data repository with news content and social context
• [MM-COVID]: COVID-19 fake news detection
• [Fin-Fact]: Fact-checking in financial domain
• [UPFD]: Graph-based fake news detection in social media

Project Members

• Kai Shu (Professor, PI)
• Canyu Chen (PhD Student)
• Haoran Wang (PhD Student)
• Aman Rangapur (Master Student)

Acknowledgments

Created by Kai Shu who can be reached at kai.shu at emory.edu.

Last Upadted: July 30, 2024