Publications

Lukas Fluri, Leon Lang, Alessandro Abate, Patrick Forré, David Krueger, Joar Skalse (2025). The Perils of Optimizing Learned Reward Functions: Low Training Error Does Not Guarantee Low Regret. ICML 2025.
PDF Cite
Lukas Fluri, Daniel Paleka, Florian Tramèr (2024). Evaluating Superhuman Models with Consistency Checks. SaTML 2024.
PDF Cite Code