Out-of-Distribution Robust Explainer for Graph Neural Networks
Experiments on synthetic and real-world datasets show that ORExplainer consistently provides more robust explanations across diverse node-level OOD settings, outperforming existing baselines.