EPFL Computer and Communication Sciences

💫 Follow us on social media! 💫 The EPFL School of Computer and Communication Sciences (IC) will no longer actively post on X (formerly Twitter). You can continue following us on platforms that better reflect how we share and connect. 👉 You can now find us on: 🌐 Mastodon (@icepfl@social.epfl.ch) 🌐 BlueSky (@icepfl.bsky.social) 📸 Instagram (@icepfl) 💼 LinkedIn (@icepfl) Each platform has its own style… it’s up to you to choose (or follow them all ) ! 📸 Instagram Dive into the life of our community through events, latest news, photos and videos. 💼 LinkedIn Key highlights, major events and collaborations across EPFL. 🌐 BlueSky & Mastodon Stay up to date with news, announcements and campus life in a direct and informative format. 👉 Follow us so you don’t miss anything!

🏆#EPFL #LakeCTF 2026 — it's a wrap! 🚀 Another year, another amazing battle of brains of creativity!🧠💻 Congratulations to this year's winning team pls carry :3! And thank you to every participant for making this edition such a success.👏 🥇 pls carry :3 🥈 .;.;. 🥉 FluxFingers

🚀 LakeCTF is LIVE! All challenges dropped at once—teams are already deep in it ⚡ Behind-the-scenes setup 👇more updates later 👀 #LakeCTF #CTF #CyberSecurity

💥Our challenge authors are preparing some interesting challenges for #LakeCTF, and we’re working hard to playtest them all. We can’t wait to see our participants tackle what we’ve cooked up! Let's go!🤘🏽 @asymmetric_re @zellic_io @infomaniak @orangecyberch @osec_io #CTF #EPFL

🚀The LakeCTF Finals coming to @EPFL_en on May 1st! Top 10 teams. One unforgettable showdown in Lausanne. 🥳 Web, crypto, reverse and more - LakeCTF is where student cybersecurity talent meets competition, collaboration, and community. More info 👉🏽lakectf.epfl.ch

@splacour40 The discussion explores the ethical challenges of AI-assisted decision-making, its environmental footprint, and whether the benefits truly outweigh the costs.

Followed by a round-table with Aimee van Wynsberghe, Katharina Frey, Executive Director of ICAIN (ETHZ), Silvia Quarteroni, Chief Transformation Officer and Head of Innovation at the Swiss Data Science Center (SDSC, EPFL) and Rachid Guerraoui, professor at Distributed Computing Laboratory at EPFL, moderated by Julie Winz.

Stéphanie Lacour @splacour40, VP for Support to Strategic Initiatives at EPFL, opens the conference “When AI Meets Sustainability: Environmental, Social and Ethical Perspectives” with welcome remarks.

It’s not too late to sign up for the workshops for tomorrow during the Sustainability Days 2026! Register here: forms.office.com/pages/response…

Professor Charlotte Bunne now presents AI in precision oncology.

Now continuing with the second session on Health innovation 👩‍⚕️🩻

Followed by Prof. Antoine Bosselut, presenting Apertus, a fully open, transparent and multilingual language model. #Apertus

The first session about Digital & AI is starting now with Prof. Sabine Süsstrunk.

The Engineering Industry Day starts with the welcome & opening addresses of Prof. Anna Fontcuberta i Morral. #EngineeringIndustryDay

The IC Faculty is present today at the Engineering Industry day with our professors and a few stands

📣💫 The 5th episode of #AIUnplugged is available on our YouTube channel ! The series explores how #AI is already being used across industries globally – in business, research, humanitarian action, government, and planning. 🔎🛡This week, we’re diving into AI and #cybersecurity with @KathrinGrosse , research scientist at @IBMResearch Zurich. Her work focuses on bridging the gap between AI #SecurityResearch and the real-world problems that AI users face every day. youtube.com/watch?v=9owXkv…

Studies show that RL sharpens learned skills and can enable real generalization, but only with the right setup: 🏆 strong verification, dense rewards, structured training, and challenging data.

And in this, what role does RL play? 👩‍🎓 In a world where models have seen almost everything, measuring real generalization is hard, so she tested reasoning across 3 levels:
→ applying a learned algorithm
→ combining multiple learned ones
→ inventing entirely new solutions

AI can solve Olympiad math… 🧮 
but still fails simple reasoning tasks 🤯 Today’s IC Colloquium by Nouha Dziri explores why and how we can build models that think more reliably 🤓.