Addie Foote

2 accepted papers

2026

Modular Pretraining Enables Access Control

Ethan Roland, Murat Cubuktepe, Erick Martinez, Stijn Servaes, Keenan Pepper, Michael Vaiana +5

ICML 2026spotlight

AI developers face a dual-use dilemma. The same capability that helps one user cure a disease can help another synthesize one. This dilemma could be resolved by access control, granting different users access to different AI capabilities. A gold standard for access control would be to serve models w…

Cited by 0 Source ↗Scholar ↗

2025

Distillation Robustifies Unlearning

Bruce W. Lee, Addie Foote, Alex Infanger, Leni Shor, Harish K Kamath, Jacob Goldman-Wetzler +3

NeurIPS 2025spotlight

Current LLM unlearning methods are not robust. A few steps of finetuning can revert their effects. We begin by showing that this is true even for an idealized form of unlearning: training to imitate a model that was never trained on unwanted information. This shows that training a model can drastica…

Cited by 0 Source ↗Scholar ↗