The field of fake news detection and graph adversarial resilience is moving towards developing more robust and innovative methods to combat the spread of misinformation. Researchers are exploring new approaches, such as using physiological signals and multi-platform propagation analysis, to improve the accuracy of fake news detection. Additionally, there is a growing focus on evaluating the robustness of graph-based detectors and developing strategies to restore fragmented social networks after attacks. Noteworthy papers include:

• A novel Structural Information principles-guided Adversarial Attack Framework, which significantly outperforms state-of-the-art baselines in attack effectiveness and enhances GNN-based detection robustness.
• A generalized theoretical framework to show the existence of critical adversarial resilience state, which can significantly outperform the state-of-the-art defense methods under various real-world datasets and attacks.
• A multi-platform fake news detection model that uses graph neural networks to extract social context features from various platforms, improving fake news detection performance by accounting for cross-platform propagation differences.