arXiv:2607. 16348v1 Announce Type: cross Abstract: Machine learning-based intrusion detection systems (IDSs) often suffer from class imbalance and vulnerability to adversarial attacks, leading to degraded detection performance and reduced robustness.
arXiv:2607. 01305v1 Announce Type: cross Abstract: Intrusion Detection Systems (IDSs) are essential for monitoring network traffic and identifying malicious activities in modern cyber-physical, Internet of Things (IoT), enterprise, and distributed network environments.
The paper introduces GraphGAN, a Graph-based Generative Adversarial Network designed to detect Distributed Denial-of-Service (DDoS) attacks in next-generation networks. It transforms sequential traffic flows into k‑nearest neighbor graphs, uses a generator to create realistic minority samples, and employs Graph Convolutional Networks for both discrimination and final classification. Experiments on four benchmark datasets demonstrate that GraphGAN outperforms existing methods in accuracy, precision, and recall, especially when data are scarce.
arXiv:2606. 28439v1 Announce Type: cross Abstract: Deep neural networks (DNNs) are widely applied in Network-based Intrusion Detection System (NIDS) due to their high accuracy.
arXiv:2603. 17717v4 Announce Type: replace-cross Abstract: Supervised detection of network attacks has always been a critical part of network intrusion detection systems (NIDS).
arXiv:2607. 17105v1 Announce Type: cross Abstract: It is crucial to safeguard computer networks from evolving network security threats and unknown cyberattacks.
Diff‑DDoS is a three‑phase framework that uses tabular diffusion models to generate realistic cyber‑physical attacks and strengthen DDoS detectors for 5G‑enabled systems. First, a CNN cell‑level detector is trained on call detail record (CDR) grids; second, a tabular denoising diffusion probabilistic model (TabDDPM) learns normal CDR aggregates to synthesize realistic attacks; third, adversarial diffusion training (ADT) iteratively produces hard, distribution‑preserving samples that harden the detector. On the Milano CDR dataset, ResNet50 with ADT achieves near‑perfect F1‑scores across multiple attack scenarios, outperforming existing synthetic‑data methods such as CTGAN.
arXiv:2503. 17867v3 Announce Type: replace-cross Abstract: Distributed Denial of Service attacks represent an active cybersecurity research problem.
arXiv:2607. 04145v1 Announce Type: new Abstract: Adversarial attacks guide and provide additional training and test data for both adversarial training and adversarial robustness validation, and expose the 'piecewise linearity' of deep learning based models.
arXiv:2606. 12075v1 Announce Type: cross Abstract: Network Intrusion Detection Systems (NIDS) heavily utlize Machine Learning (ML) but ML models can be manipulated via adversarial attacks.
arXiv:2511. 13749v2 Announce Type: replace Abstract: Deep neural networks are known to be vulnerable to adversarial perturbations, which are small, carefully crafted inputs that lead to incorrect predictions.
arXiv:2606. 01437v1 Announce Type: cross Abstract: Deep Neural Networks (DNNs) are highly susceptible to adversarial perturbations, leading to extensive research on robustness for safety-critical applications.