Publications

Luca Mautino, Mirza Mujtaba Hussain, Iacopo Masi. Correlating Adversarial Robustness with the Energy of Untrained Classifiers. International Conference on Machine Learning and Cybernetics (ICMLC) best-paper finalist 🥈, 2026.

Giuseppe Daidone, Filippo Bartolucci, Maria Rosaria Briglia, Mirza Mujtaba Hussain, Giuseppe Lisanti, Iacopo Masi. Neutralizing Proactive Defense using Diffusion-based Upsampling. ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec), 2026.

Abzal Aidakhmetov, Donato Crisostomi, Tommaso Mencattini, Robert Adrian Minut, Iacopo Masi, Emanuele Rodolà. Steering Vectors are an Adversarial Attack Surface. Empirical Methods in Natural Language Processing (EMNLP) Findings, 2026.

Mario Iacobelli, Robert Adrian Minut, Tommaso Mencattini, Donato Crisostomi, Andrea Santilli, Iacopo Masi, Emanuele Rodolà. Multi-objective Evolutionary Merging Enables Efficient Reasoning Models. Conference on Language Modeling (COLM), 2026.

Maria Rosaria Briglia, Simone Facchiano, Paolo Cursi, Alessio Sampieri, Emanuele Rodolà, Guido Maria D'Amely Di Melendugno, Luca Franco, Fabio Galasso, Iacopo Masi. Not All Latent Spaces Are Flat: Hyperbolic Concept Control. arXiv preprint, 2026.

Mirza Mujtaba Hussain, Antonio D'Orazio, Odelia Melamed, Iacopo Masi. A Provable Energy-Guided Test-Time Defense Boosting Adversarial Robustness of Large Vision-Language Models. IEEE/CVF Computer Vision and Pattern Recognition (CVPR), 2026.

Maria Rosaria Briglia, Mirza Mujtaba Hussain, Giuseppe Lisanti, Iacopo Masi. Why Adversarially Train Diffusion Models?. International Conference on Learning Representations (ICLR), 2026.

Robert Adrian Minut, Hazem Dewidar, Iacopo Masi. Spilled Energy in Large Language Models. International Conference on Learning Representations (ICLR), 2026.

Antonio D'Orazio, Maria Rosaria Briglia, Donato Crisostomi, Dario Loi, Emanuele Rodolà, Iacopo Masi. Implicit Inversion turns CLIP into a Decoder. International Conference on Learning Representations (ICLR), 2026.

Igor Maljkovic, Maria Rosaria Briglia, Iacopo Masi, Antonio Emanuele Cinà, Fabio Roli. Harnessing Hyperbolic Geometry for Harmful Prompt Detection and Sanitization. International Conference on Learning Representations (ICLR), 2026.

Donato Crisostomi, Alessandro Zirilli, Antonio Andrea Gargiulo, Maria Sofia Bucarelli, Simone Scardapane, Fabrizio Silvestri, Iacopo Masi, Emanuele Rodolà. MASS : MoErging through Adaptive Subspace Selection. International Conference on Learning Representations (ICLR), 2026.

Davide Gabrielli, Simone Sestito, Iacopo Masi. Inverse Language Modeling for Robust and Grounded LLMs. AAAI 2026 Workshop on Language Models for Underserved Communities best-poster runner up award 🥈, 2025.

Kartik Thakral, Shreyansh Pathak, Tamar Glaser, Tal Hassner, Diego Garcia-Olano, Iacopo Masi, Richa Singh, Mayank Vatsa. Genμ: The Generative Machine Unlearning Challenge. IEEE/CVF ICCV Workshop on Unlearning and Model Editing U&ME, 2025.

Mirza Mujtaba Hussain, Maria Rosaria Briglia, Filippo Bartolucci, Senad Beadini, Giuseppe Lisanti, Iacopo Masi. Understanding Adversarial Training with Energy-based Models. arXiv preprint (technical report), 2025.

Filippo Bartolucci, Iacopo Masi, Giuseppe Lisanti. Perturb, Attend, Detect, and Localize (PADL): Robust Proactive Image Defense. IEEE Access, 2025.

Xiao Guo, Xiaohong Liu, Iacopo Masi, Xiaoming Liu. Language-guided Hierarchical Fine-grained Image Forgery Detection and Localization. International Journal of Computer Vision (IJCV), 2024.

Antonio D'Orazio, Davide Sforza, Fabio Pellacini, Iacopo Masi. Environment Maps Editing using Inverse Rendering and Adversarial Implicit Functions. Smart Tools and Applications in Graphics (STAG), 2024.

Mirza Mujtaba Hussain, Maria Rosaria Briglia, Senad Beadini, Iacopo Masi. Shedding More Light on Robust Classifiers Under the Lens of Energy-Based Models. European Conference on Computer Vision (ECCV), 2024.

Xiao Guo, Xiaohong Liu, Zhiyuan Ren, Steven Grosz, Iacopo Masi, Xiaoming Liu. Hierarchical Fine-grained Image Forgery Detection and Localization. Conference on Computer Vision and Pattern Recognition (CVPR), 2023.

Senad Beadini, Iacopo Masi. Exploring the Connection between Robust and Generative Models. Italian Conference on AI - Ital-IA - Workshop on AI for Cybersecurity, 2023.

Mozhdeh Rouhsedaghat, Masoud Monajatipoor, Kai-Wei Chang, C-C Jay Kuo, Iacopo Masi. MAGIC: Mask-Guided Image Synthesis by Inverting a Quasi-Robust Classifier. Conference on Artificial Intelligence (AAAI), 2022.

Y Nirkin, I Masi, Anh Tuan Tran, T Hassner, G Medioni. Face Segmentation, Face Swapping, and How They Impact Face Recognition. Book Chapter in Deep Learning-Based Face Analytics, 2021.

Dorjan Hitaj, Giulio Pagnotta, Iacopo Masi, Luigi v Mancini. Evaluating the Robustness of Geometry-Aware Instance-Reweighted Adversarial Training. arXiv preprint (technical report), 2021.