Is Multi-Modal Necessarily Better? Robustness Evaluation of Multi-Modal Fake News Detection
Open accessAbstract
The proliferation of fake news and its serious negative social influence push fake news detection methods to become necessary tools for web managers. Meanwhile, the multi-media nature of social media makes multi-modal fake news detection popular for its ability to capture more modal features than uni-modal detection methods. However, current literature on multi-modal detection is more likely to pursue the detection accuracy but ignore the robustness (the detection ability in the case of abnormality and malicious attack) of the detector. To address this problem, we propose a comprehensive robustness evaluation of multi-modal fake news detectors. In this work, we simulate the attack methods of malicious users and developers, i.e., posting fake news and injecting backdoors. Specifically, we evaluate multi-modal detectors with five adversarial and two backdoor attack methods. Experiment results imply that: (1) The detection performance of the state-of-the-art detectors degrades significantly under adversarial attacks, e.g., BDANN's detection accuracy on malicious news drops by 47% compared to normal, even worse than general detectors (Att-RNN); (2) Most multimodal detectors are more vulnerable to visual modality than textual modality; (3) Backdoor attacks on popular events news severely degrade detectors (accuracy dropped by an average of 20%); (4) These detectors degrade more (another 2% reduction in accuracy) when subjected to multi-modal attacks; (5) Defense methods will improve the robustness of multi-modal detectors, but cannot fully resist the effects of malicious attacks.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Multimodal models may be equally or even more vulnerable to adversarial attacks than models using a single modality, due to negative impact of increasing input dimensions on adversarial robustness.
“The robustness of multimodal models is a topic of ongoing discussion because multimodal models may be equally or even more vulnerable to adversarial attacks than models using a single modality. This susceptibility to adversarial attacks results from the negative impact of increasing input dimensions on adversarial robustness49–51.”
From Orchestrating explainable artificial intelligence for multimodal and longitudinal data in medical imaging (2024) · cites this paper as [51]
Cites
(2008) · Journal of Machine Learning Research · Cited by 33,051 · see citing works
Rethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138 · see citing works
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381 · see citing works
Towards Evaluating the Robustness of Neural Networks
Carlini, N., & Wagner, D. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.49 · Cited by 8,859 · see citing works
DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks
Moosavi-Dezfooli, S.-M., Fawzi, A., & Frossard, P. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.282 · Cited by 5,317 · see citing works
The Limitations of Deep Learning in Adversarial Settings
Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z. B., & Swami, A. (2016) · 2016 IEEE European Symposium on Security and Privacy (EuroS&P) · doi:10.1109/eurosp.2016.36 · Cited by 3,993 · see citing works
Fake News Detection on Social Media: A Data Mining Perspective
Shu, K., Sliva, A., Wang, S., Tang, J., & Liu, H. (2017) · ACM SIGKDD Explorations Newsletter · doi:10.1145/3137597.3137600 · Cited by 3,304 · see citing works
How to backdoor federated learning
Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2018) · arXiv · doi:10.48550/arxiv.1807.00459 · Cited by 1,495 · see citing works
Adversarial Examples for Evaluating Reading Comprehension Systems
Jia, R., & Liang, P. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1215 · Cited by 1,410 · see citing works
Trojaning Attack on Neural Networks
Liu, Y., Ma, S., Aafer, Y., Lee, W.-C., Zhai, J., Wang, W., & Zhang, X. (2018) · Proceedings 2018 Network and Distributed System Security Symposium · doi:10.14722/ndss.2018.23291 · Cited by 1,270 · see citing works
BadNets: Evaluating Backdooring Attacks on Deep Neural Networks
Gu, T., Liu, K., Dolan-Gavitt, B., & Garg, S. (2019) · IEEE Access · doi:10.1109/access.2019.2909068 · Cited by 1,238 · see citing works
Wang, Y., Ma, F., Jin, Z., Yuan, Y., Xun, G., Jha, K., Su, L., & Gao, J. (2018) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3219819.3219903 · Cited by 1,181 · see citing works
Jin, D., Jin, Z., Zhou, J. T., & Szolovits, P. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6311 · Cited by 1,108 · see citing works
HotFlip: White-Box Adversarial Examples for Text Classification
Ebrahimi, J., Rao, A., Lowd, D., & Dou, D. (2018) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p18-2006 · Cited by 948 · see citing works
Multimodal Fusion with Recurrent Neural Networks for Rumor Detection on Microblogs
Jin, Z., Cao, J., Guo, H., Zhang, Y., & Luo, J. (2017) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3123266.3123454 · Cited by 859 · see citing works
MVAE: Multimodal Variational Autoencoder for Fake News Detection
Khattar, D., Goud, J. S., Gupta, M., & Varma, V. (2019) · International World Wide Web Conference (WWW) · doi:10.1145/3308558.3313552 · Cited by 848 · see citing works
Attention-Based Recurrent Neural Network Models for Joint Intent Detection and Slot Filling
Liu, B., & Lane, I. R. (2016) · Interspeech (USB) · doi:10.21437/interspeech.2016-1352 · Cited by 723 · see citing works
A Stylometric Inquiry into Hyperpartisan and Fake News
Potthast, M., Kiesel, J., Reinartz, K., Bevendorff, J., & Stein, B. (2018) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p18-1022 · Cited by 700 · see citing works
Generating Natural Language Adversarial Examples through Probability Weighted Word Saliency
Ren, S., Deng, Y., He, K., & Che, W. (2019) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p19-1103 · Cited by 666 · see citing works
Inoculating Against Fake News About COVID-19
van der Linden, S., Roozenbeek, J., & Compton, J. (2020) · Frontiers in Psychology · doi:10.3389/fpsyg.2020.566790 · Cited by 632 · cited 1× · see citing works
“Moreover, during the outbreak of the COVID-19, lots of fake news about this pandemic on social media have harmed people’s health-protective behaviors [2].”
BackgroundSpotFake: A Multi-modal Framework for Fake News Detection
Singhal, S., Shah, R. R., Chakraborty, T., Kumaraguru, P., & Satoh, S. (2019) · IEEE International Conference on Multimedia Big Data · doi:10.1109/bigmm.2019.00-44 · Cited by 591 · see citing works
TextAttack: A Framework for Adversarial Attacks, Data Augmentation, and Adversarial Training in NLP
Morris, J. X., Lifland, E., Yoo, J. Y., Grigsby, J., Jin, D., & Qi, Y. (2020) · Conference on Empirical Methods in Natural Language Processing: System Demonstrations (EMNLP) · doi:10.18653/v1/2020.emnlp-demos.16 · Cited by 570 · see citing works
Meel, P., & Vishwakarma, D. K. (2019) · Expert Systems with Applications · doi:10.1016/j.eswa.2019.112986 · Cited by 529 · see citing works
Novel Visual and Statistical Image Features for Microblogs News Verification
Jin, Z., Cao, J., Zhang, Y., Zhou, J., & Tian, Q. (2016) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2016.2617078 · Cited by 470 · see citing works
Fighting Fake News: Image Splice Detection via Learned Self-Consistency
Huh, M., Liu, A., Owens, A., & Efros, A. A. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01252-6_7 · Cited by 465 · see citing works
Sahoo, S. R., & Gupta, B. B. (2020) · Applied Soft Computing · doi:10.1016/j.asoc.2020.106983 · Cited by 445 · see citing works
Weight Poisoning Attacks on Pretrained Models
Kurita, K., Michel, P., & Neubig, G. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.249 · Cited by 354 · see citing works
Can You Really Backdoor Federated Learning?
Sun, Z., Kairouz, P., Suresh, A. T., & McMahan, H. B. (2019) · arXiv · doi:10.48550/arxiv.1911.07963 · Cited by 348 · see citing works
Disinformation on the Web: Impact, Characteristics, and Detection of Wikipedia Hoaxes
Kumar, S., West, R., & Leskovec, J. (2016) · Proceedings of the 25th International Conference on World Wide Web · doi:10.1145/2872427.2883085 · Cited by 310 · see citing works
A Backdoor Attack Against LSTM-Based Text Classification Systems
Dai, J., Chen, C., & Li, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2941376 · Cited by 298 · see citing works
Fairness in Deep Learning: A Computational Perspective
Du, M., Yang, F., Zou, N., & Hu, X. (2020) · IEEE Intelligent Systems · doi:10.1109/mis.2020.3000681 · Cited by 218 · see citing works
A Review on Explainability in Multimodal Deep Neural Nets
Joshi, G., Walambe, R., & Kotecha, K. (2021) · IEEE Access · doi:10.1109/access.2021.3070212 · Cited by 204 · see citing works
CED: Credible Early Detection of Social Media Rumors
Song, C., Yang, C. H., Chen, H., Tu, C., Liu, Z., & Sun, M. (2019) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2019.2961675 · Cited by 199 · see citing works
Defensive Modeling of Fake News Through Online Social Networks
Shrivastava, G., Kumar, P., Ojha, R. P., Srivastava, P. K., Mohan, S., & Srivastava, G. (2020) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2020.3014135 · Cited by 158 · see citing works
SAME: Sentiment-Aware Multi-Modal Embedding for Detecting Fake News
Cui, L.-M., Wang, S., & Lee, D. (2019) · International Conference on Advances in Social Networks Analysis and Mining · doi:10.1145/3341161.3342894 · Cited by 142 · see citing works
BDANN: BERT-Based Domain Adaptation Neural Network for Multi-Modal Fake News Detection
Zhang, T., Wang, D., Chen, H., Zeng, Z., Guo, W., Miao, C., & Cui, L. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9206973 · Cited by 140 · see citing works
Verifying Multimedia Use at MediaEval 2015
Boididou, C., Papadopoulos, S., Dang‐Nguyen, D., Boato, G., Riegler, M. A., Middleton, S. E., Petlund, A., & Kompatsiaris, I. (2015) · Iris (University of Trento) · Cited by 136 · see citing works
Fake News Mitigation via Point Process Based InterventionPreprint
Farajtabar, M., Yang, J., Ye, X., Xu, H., Trivedi, R., Khalil, E. B., Li, S., Song, L., & Hongyuan Zha, B. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.07823 · Cited by 121 · see citing works
HAN, image captioning, and forensics ensemble multimodal fake news detection
Meel, P., & Vishwakarma, D. K. (2021) · Information Sciences · doi:10.1016/j.ins.2021.03.037 · Cited by 113 · see citing works
Tembhurne, J. V., Almin, M. M., & Diwan, T. (2022) · International Journal on Semantic Web and Information Systems · doi:10.4018/ijswis.295553 · Cited by 109 · see citing works
Detection and veracity analysis of fake news via scrapping and authenticating the web search
Vishwakarma, D. K., Varshney, D., & Yadav, A. (2019) · Cognitive Systems Research · doi:10.1016/j.cogsys.2019.07.004 · Cited by 108 · see citing works
Knowledge-aware Multi-modal Adaptive Graph Convolutional Networks for Fake News Detection
Qian, S., Hu, J., Fang, Q., & Xu, C. (2021) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3451215 · Cited by 100 · see citing works
A temporal ensembling based semi-supervised ConvNet for the detection of fake news articles
Meel, P., & Vishwakarma, D. K. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.115002 · Cited by 85 · see citing works
A review on rumour prediction and veracity assessment in online social network
Varshney, D., & Vishwakarma, D. K. (2020) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.114208 · Cited by 82 · see citing works
An empirical approach to understanding users' fake news identification on social media
Aoun Barakat, K., Dabbous, A., & Tarhini, A. (2021) · Online Information Review · doi:10.1108/oir-08-2020-0333 · Cited by 80 · see citing works
A unified approach for detection of Clickbait videos on YouTube using cognitive evidences
Varshney, D., & Vishwakarma, D. K. (2021) · Applied Intelligence · doi:10.1007/s10489-020-02057-9 · Cited by 73 · see citing works
Mishra, A., Gupta, B. B., & Joshi, R. C. (2011) · European Intelligence and Security Informatics Conference · doi:10.1109/eisic.2011.15 · Cited by 70 · see citing works
Zheng, H., Chen, Z., Du, T., Zhang, X., Cheng, Y., Ji, S., Wang, J., Yu, Y., & Chen, J. (2022) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1145/3510003.3510123 · Cited by 63 · see citing works
Self-citationGupta, B. B., Gupta, S. R., Gangwar, S. K., et al. (2015) · Journal of Information Privacy and Security · doi:10.1080/15536548.2015.1044865 · Cited by 62 · see citing works
Ali, H. M., Khan, M. S., AlGhadhban, A., Alazmi, M., Alzamil, A., Al-Utaibi, K. A., & Qadir, J. (2021) · IEEE Access · doi:10.1109/access.2021.3085875 · Cited by 59 · see citing works
ConvNet frameworks for multi-modal fake news detection
Raj, C., & Meel, P. (2021) · Applied Intelligence · doi:10.1007/s10489-021-02345-y · Cited by 53 · see citing works
Text Processing Like Humans Do: Visually Attacking and Shielding NLP SystemsPreprint
Eger, S., Şahin, G. G., Rücklé, A., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.11508 · Cited by 49 · see citing works
On Modality Bias Recognition and Reduction
Guo, Y., Nie, L., Cheng, H. H., Cheng, Z., Kankanhalli, M., & Bimbo, A. D. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3565266 · Cited by 48 · see citing works
Varshney, D., & Vishwakarma, D. K. (2020) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-020-02698-1 · Cited by 41 · see citing works
Recent State-of-the-art of Fake News Detection: A Review
Vishwakarma, D. K., & Jain, C. (2020) · International Conference for Emerging Technology (INCET) · doi:10.1109/incet49848.2020.9153985 · Cited by 37 · see citing works
Malhotra, B., & Vishwakarma, D. K. (2020) · IEEE International Conference on Multimedia Big Data · doi:10.1109/bigmm50055.2020.00034 · Cited by 34 · see citing works
GRIP-GAN: An Attack-Free Defense Through General Robust Inverse Perturbation
Zheng, H., Chen, J., Du, H., Zhu, W., Ji, S., & Zhang, X. (2021) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2021.3124337 · Cited by 28 · see citing works
Self-citationMachine Learned Classifiers for Trustworthiness Assessment of Web Information Contents
Meel, P., & Vishwakarma, D. K. (2021) · International Conference on Computing, Communication, and Intelligent Systems (ICCCIS) · doi:10.1109/icccis51004.2021.9397228 · Cited by 10 · see citing works
Towards Evaluating the Robustness of Chinese BERT ClassifiersPreprint
Wang, B., Pan, B., Li, X. Y., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.03742 · Cited by 8 · see citing works
Deep Neural Architecture for Veracity Analysis of Multimodal Online Information
Meel, P., & Vishwakarma, D. K. (2021) · International Conference on Cloud Computing, Data Science & Engineering (Confluence) · doi:10.1109/confluence51648.2021.9377172 · Cited by 7 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Pahud de Mortanges, A., Luo, H., Shu, S. Z., Kamath, A., Suter, Y., Shelan, M., Pöllinger, A., & Reyes, M. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01190-w · Cited by 93 · see citing works
“This susceptibility to adversarial 49 – 51 .”
- Authors
- Chen, J., Jia, C., Zheng, H., Chen, R., & Fu, C.
- Affiliations
- Zhejiang University of Technology (CN)
- Venue
- IEEE Transactions on Network Science and Engineering
- Published
- 2023
- Pages
- 1-15
- ISSN
- 2327-4697 (ISSN-L)
- 2334-329X
- OpenAlex
- W4322576826
- Cited by
- 17
- References
- 60
- Citations by year
- 2026: 2 · 2025: 7 · 2024: 5 · 2023: 3
- FWCI
- 7.34
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Network Science and Engineering (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (62072406, 2021C01117)
- National Key Laboratory of Science and Technology on Information System Security (61421110502)
- Zhejiang Provincial Natural Science Foundation (LDQ23F020001)
- Key R&D Projects in Zhejiang
- National Key R&D Projects of China (2018AAA0100801)
- Topics
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- fake news detection, multimodal detection, robustness analysis, adversarial attacks, backdoor attacks, social media misinformation, defense methods
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
