🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

After many requests, we have extended abstract/paper submission deadlines by a few days ! Abstract by March 3 and full papers by March 6. More info here: worma.gitlab.io/2025/

It’s always exciting to see your research making an impact and being referred by industry stakeholders. Our “Hi Mum and Dad” SMS scam paper set to appear USENIX Security, has been referred by Phonely in their latest blog article on #himum #scams! phonely.co.uk/blogs/guides/t…

On March 6th, 2025 (6pm-8pm UK time), there will be a PhD Open Evening, as an hybrid event held **online via Zoom** and in person at "Room 225, Central House," 14 Upper Woburn Place, WC1H 0NN, London. You may register here: eventbrite.co.uk/e/phd-hybrid-o… /cc UCL InfoSec UCL Computer Science

🚨 Fully Funded PhD Opportunity at UCL! 🔍 Enhancing Crime Detection in the IoT Era 💰 Covers: UK fees, stipend, project costs 📅 Deadline: Apr 15 2025 👨‍🏫 Supervisors: Dr. Mandalari, Dr. Nilufer Tuptuk, Dr. Fabio Pierazzi More info: [email protected], ucl.ac.uk/security-crime…

The SaTML Conference program is out and our paper "ML-Based Behavioral Malware Detection Is Far From a Solved Problem" is there! Check preprint: arxiv.org/abs/2405.06124 It was amazing to work with great researchers on it Yigitcan Kaya Fabio Pierazzi Lorenzo Cavallaro

I’m excited to share our latest work at #SaTML2025, "ML-Based Behavioral Malware Detection Is Far from a Solved Problem"! Joint work with Yizheng Chen, Marcus Botacin, Shoumik Saha, Fabio Pierazzi, Lorenzo Cavallaro, David Wagner, Tudor Dumitraș 🔗Website: malwaredetectioninthewild.github.io 👇(1/5)

(2/5):🔍We expose that ML-based malware detectors trained on sandbox data fall short when deployed on real-world endpoints. Testing on sandbox traces: ~95% recall (prior work) Testing on real-world endpoint traces: <20% recall (our work) That’s a 75% drop in effectiveness!

(3/5):🔬Why does this happen? Sandbox traces lack real-world variability—models learn spurious features that don’t generalize. Endpoint detectors only see the hard-to-classify samples—those that can bypass static defenses. Result? Endpoint malware detection is an open problem.

(4/5): 📢 We call the community to action and provide: A dataset of sandbox traces for model development A community evaluation pipeline that tests submitted models on real endpoint data. A leaderboard to rank submissions & highlight approaches. 🔗Visit malwaredetectioninthewild.github.io

(5/5): 🛡️ Why this matters: ML models trained on public datasets might not generalize to real threats. Security companies hold valuable data but can’t share it. 🏆 Our leaderboard provides a blueprint for evaluating models realistically—without exposing sensitive data.

🚨 Call for Papers: CAMLIS 2025 is officially OPEN! 🎉 camlis.org/call-for-papers PLUS, Student Scholarships are now available! 🍎 Apply today: forms.gle/keLpB9LLAj2QRy…

I have an open PhD position at UCL InfoSec on applying traffic-analysis resistance techniques to protect industrial control systems. Full funding is available for home-fee status students (deadline 15 April). ucl.ac.uk/security-crime…

A few days left for this PhD opportunity (application due April 15th). Details below! UCL InfoSec UCL Computer Science

🚀 Exciting Announcement! 🚀 Get ready for the 18th ACM Workshop on Artificial Intelligence and Security (AISec 2025)! 📍Co-located: ACM CCS 2024 🗓️ Deadline: June 20th, 2025 🌐 Website: aisec.cc w/ Ruoxi Jia and Matthew Jagielski

Join the Graz Security Week from Sep 1 to 5! with Sahar Abdelnabi 🕊 (on 🦋), Jo Van Bulck, Maria Eichlseder, Georg Fuchsbauer, David Oswald, Fabio Pierazzi, Kav, Christian Rossow (@[email protected]), Yang Zhang on topics system security, side channels, AI Security, and Cryptography: securityweek.at