Publications

Selected scientific publications

2026

  1. SOM Directions are Better than One: Multi-Directional Refusal Suppression in Language Models
    Giorgio Piras, Raffaele Mura, Fabio Brau, and 3 more authors
    arXiv preprint arXiv:2511.08379, 2026

2025

  1. Video Deblurring by Sharpness Prior Detection and Edge Information
    Yang Tian, Fabio Brau, Giulio Rossolini, and 2 more authors
    arXiv preprint arXiv:2501.12246, 2025
  2. S2AP: Score-space Sharpness Minimization for Adversarial Pruning
    Giorgio Piras, Qi Zhao, Fabio Brau, and 3 more authors
    arXiv preprint arXiv:2510.18381, 2025
  3. TransferBench: Benchmarking Ensemble-based Black-box Transfer Attacks
    Fabio Brau, Maura Pintor, Antonio Emanuele Cinà, and 5 more authors
    Advances in Neural Information Processing Systems, 2025
  4. rossolini2025exploiting.png
    Exploiting edge features for transferable adversarial attacks in distributed machine learning
    Giulio Rossolini, Fabio Brau, Alessandro Biondi, and 2 more authors
    Internet of Things, 2025

2024

  1. PhD Thesis
    brau2024methods.png
    Methods for Certifiable Robustness of Deep Neural Networks
    Fabio Brau
    Mar 2024
  2. 1-Lipschitz Layers Compared: Memory Speed and Certifiable Robustness
    Bernd Prach, Fabio Brau, Giorgio Buttazzo, and 1 more author
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2024

2023

  1. Defending from Physically-Realizable Adversarial Attacks through Internal Over-Activation Analysis
    Giulio Rossolini, Federico Nesti, Fabio Brau, and 2 more authors
    Proceedings of the AAAI Conference on Artificial Intelligence, Jun 2023
  2. Robust-by-Design Classification via Unitary-Gradient Neural Networks
    Fabio Brau, Giulio Rossolini, Alessandro Biondi, and 1 more author
    Proceedings of the AAAI Conference on Artificial Intelligence, Jun 2023
  3. bisconti2023maximizing.png
    Maximizing team synergy in AI-related interdisciplinary groups: an interdisciplinary-by-design iterative methodology
    Piercosma Bisconti, Davide Orsitto, Federica Fedorczyk, and 8 more authors
    AI & SOCIETY, Jun 2023
  4. On the Minimal Adversarial Perturbation for Deep Neural Networks With Provable Estimation Error
    Fabio Brau, Giulio Rossolini, Alessandro Biondi, and 1 more author
    IEEE Transactions on Pattern Analysis and Machine Intelligence, Jun 2023

2021

  1. lanciano2021using.png
    Using Self-Organizing Maps for the Behavioral Analysis of Virtualized Network Functions
    Giacomo Lanciano, Antonio Ritacco, Fabio Brau, and 5 more authors
    In Cloud Computing and Services Science, Jun 2021
  2. cucinotta2021forecasting.png
    Forecasting operation metrics for virtualized network functions
    Tommaso Cucinotta, Giacomo Lanciano, Antonio Ritacco, and 7 more authors
    In 2021 IEEE/ACM 21st International Symposium on Cluster, Cloud and Internet Computing (CCGrid), Jun 2021