SOURCE-LINKED INTELLIGENCE

Kernel-Complexity Edge Sanitization for Training-Free Defense against Structural Graph Attacks

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

Graph Neural Networks (GNNs) have achieved remarkable success across diverse applications, yet they remain highly vulnerable to adversarial attacks that maliciously perturb graph structure. Existing defenses often lack rigorous theoretical grounding, rely on attack-specific heuristics, or require costly retraining procedures such as adversarial training. To address these limitations, we propose Kernel-Complexity Edge Sanitization (KCES), a training-free and model-agnostic framework for defending against structural attacks. KCES is built upon Graph Kernel Complexity (GKC), a principled metric d

Read original source ↗ Open in workspace

recordType

paper

region

Global