Francesco Ortu
Hello, I'm Francesco, a third-year PhD student at University of Trieste and at Laboratory of Data Engineering (LADE) at AREA Science Park, under the supervision of Alberto Cazzaniga.
I'm currently working on Mechanistic Interpretability and AI Safety. My long-term goal is to help reduce the risks from advanced AI systems. Previously, I was a research intern at Max-Planck Institute for Intelligent Systems, working under the supervision of Zhijing Jin and Prof. Bernhard Schölkopf. During my PhD, I spent six months at the ELLIS Institute Tübingen, working with Maksym Andriushchenko.
I'm also a co-founder of AI Safety for Italy.
I’m also a sailor and recently I started to learn climbing. I grew up in a small town in the center of Italy, but I’m currently living in Trieste, a beautiful city on the sea in the north-east of Italy.
My email is francesco.ortu[at]phd.units.it.
News
- Aug 2026 — Our paper "Preserving Historical Truth: Detecting Historical Revisionism in Large Language Models" has been accepted to the EMNLP 2026 main conference! 🎉
- Apr 2026 — Our paper "When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models" has been accepted to the ACL 2026 main conference! 🎉
- Apr 2026 — I'll be joining the ELLIS Institute Tübingen and the Max Planck Institute as a visiting PhD student to work with Maksym Andriushchenko.
- Feb 2026 — Gave a talk on "Preserving Historical Truth: Detecting Historical Revisionism in LLMs" at IASEAI 2026.
- Sept 2025 — I'll present two papers at NeurIPS 2025: one at main conference and one at the Mechanistic Interpretability workshop! 🎉
- Feb 2024 — Our paper "Competition of Mechanisms" has been accepted at ACL 2024! 🎉
Selected Publications
|
|
|