mlsanityData to 14 Oct 2026

Is Multi-Modal Necessarily Better? Robustness Evaluation of Multi-modal Fake News Detection

arXiv 2206.08788 ↗First posted 17 Jun 2022cs.AI cs.CY cs.SIcited by 2

On the map: LLM safety and security → Backdoor and data poisoning attacks

Abstract (arXiv)

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 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, even worse than general detectors; (2) Most multi-modal detectors are more vulnerable when subjected to attacks on visual modality than textual modality; (3) Popular events' images will cause significant degradation to the detectors when they are subjected to backdoor attacks; (4) The performance of these detectors under multi-modal attacks is worse than under uni-modal attacks; (5) Defensive methods will improve the robustness of the multi-modal detectors.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security. Builds on 26 of its 67 references that are in the map. First posted on arXiv 17 Jun 2022.

Builds on

26 of 67 references.

References · 26

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  3. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  4. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,318
  5. 2015The Limitations of Deep Learning in Adversarial Settingscited by 790
  6. 2018How To Backdoor Federated Learningcited by 765
  7. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
  8. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 571
  9. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
  10. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 341
  11. 2017Fake News Detection on Social Media: A Data Mining Perspectivecited by 336
  12. 2019Can You Really Backdoor Federated Learning?cited by 255
14 more
  1. 2019A backdoor attack against LSTM-based text classification systemscited by 227
  2. 2016Attention-Based Recurrent Neural Network Models for Joint Intent Detection and Slot Fillingcited by 137
  3. 2018Fighting Fake News: Image Splice Detection via Learned Self-Consistencycited by 136
  4. 2019Fairness in Deep Learning: A Computational Perspectivecited by 123
  5. 2019Text Processing Like Humans Do: Visually Attacking and Shielding NLP Systemscited by 101
  6. 2017A Stylometric Inquiry into Hyperpartisan and Fake Newscited by 82
  7. 2020Shape-Texture Debiased Neural Network Trainingcited by 76
  8. 2020Removing Bias in Multi-modal Classifiers: Regularization by Maximizing Functional Entropiescited by 57
  9. 2021A Review on Explainability in Multimodal Deep Neural Netscited by 46
  10. 2021NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identificationcited by 42
  11. 2018CED: Credible Early Detection of Social Media Rumorscited by 26
  12. 2017Fake News Mitigation via Point Process Based Interventioncited by 24
  13. 2022On Modality Bias Recognition and Reductioncited by 20
  14. 2020Towards Evaluating the Robustness of Chinese BERT Classifierscited by 2

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 0; 2025: 1.

2 papers.

  1. 2023TIJO: Trigger Inversion with Joint Optimization for Defending Multimodal Backdoored Modelscited by 12
  2. 2025A New Hybrid Intelligent Approach for Multimodal Detection of Suspected Disinformation on TikTokcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Exploring Modality Disruption in Multimodal Fake News Detection
  2. 2025Modality Interactive Mixture-of-Experts for Fake News Detection
  3. 2025MM-FusionNet: Context-Aware Dynamic Fusion for Multi-modal Fake News Detection with Large Vision-Language Models
  4. 2025Deconfounded Reasoning for Multimodal Fake News Detection via Causal Intervention
  5. 2023On the Adversarial Robustness of Multi-Modal Foundation Models
  6. 2026Head-wise Modality Specialization within MLLMs for Robust Fake News Detection under Missing Modality
  7. 2026Can Multimodal Large Language Models Generate and Detect Multimodal Social Media Fake News?
  8. 2024Fake News Detection and Manipulation Reasoning via Large Vision-Language Models