LATIDIA · Ciberseguridad

GraphIP-Bench: ¿Qué tan difícil es robar una red neuronal gráfica y podemos detenerla?

arXiv: 2605.12827v3Tipo de anuncio: reemplazar Resumen: Las redes neuronales gráficas (GNN) implementadas como servicios en la nube pueden ser *robadas * a través de * ataques de extracción de modelos *, que entrenan a un sustituto de las respuestas de consulta para reproducir las

La noticia

arXiv:2605.12827v3 Announce Type: replace Abstract: Graph neural networks (GNNs) deployed as cloud services can be *stolen* through *model-extraction attacks*, which train a surrogate from query responses to reproduce the target's behaviour, and a growing line of ownership defenses tries to prevent or trace such theft. The title of this paper asks two questions: *how hard is it to steal a GNN?*, and *can we stop it?* Prior work cannot answer either, because experiments use inconsistent datasets, threat models, and metrics. We introduce *GraphIP-Bench*, a unified benchmark which evaluates both sides under a single black-box protocol. It integrates twelve extraction attacks, twelve defenses spanning watermarking, output-perturbation, and query-pattern-detection families, ten public