Plausible Counterfactuals: Auditing Deep Learning Classifiers with Realistic Adversarial Examples | Publicación