👀🛡️ Code for the paper “Blending adversarial training and representation-conditional purification via aggregation improves adversarial robustness” by Emanuele Ballarin, Alessio Ansuini and Luca Bortolussi (2025) - View it on GitHub