Our group, OmnAI Lab at Sapienza, presented a paper at IH&MMSec 2026 in Firenze, Italy 🇮🇹✨

1 paper at IH&MMSec 2026

We presented a paper at the ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec 2026), held in Firenze, Italy (17–19 June 2026). The work systematically evaluates the robustness of recent state-of-the-art proactive image-protection schemes (PADL and DiffVax) and shows that they can be neutralized by black-box, diffusion-based upsampling attacks — exposing critical gaps between assumed and real-world threat models.

• G. Daidone, F. Bartolucci, M. R. Briglia, M. H. Mirza, G. Lisanti, I. Masi - Neutralizing Proactive Defense using Diffusion-based Upsampling, IH&MMSec 2026

A nice collaboration with the University of Bologna 🤝

Acknowledgements

Work supported by ACN (Italian National Cybersecurity Agency)