Teaching

Privacy Enhancing Technologies

Large Language Models

Information Security Lab

Student projects

If you’re a student at ETH Zürich interested in a semester project or thesis, please contact us. Please include a copy of your CV, latest transcripts, and a short description of your interests.

This is a list of previous student projects at our lab:

• Lukas Fluri. Evaluating Superhuman Models with Consistency Checks.
• Debeshee Das. Blind Baselines Beat Membership Inference Attacks for Foundation Models.
• Fredrik Nestaas. Adversarial Search Engine Optimization for Large Language Models.
• Robert Hönig. Adversarial Perturbations Cannot Reliably Protect Artists From Generative AI.
• Lorenzo Rossi. Auditing Privacyt Leakage in Large Language Models.
• Shanglun Feng. Privacy Backdoors: Stealing Data with Corrupted Pretrained Models.
• Jakub Łucki. An Adversarial Perspective on Machine Unlearning for AI Safety.
• Yize Cheng. Universal Black‑box Adversarial Examples.
• Qianjun Zheng. White‑box Membership Inference Attacks.
• Joshua Freeman. Exploring Memorization and Copyright Violation in Frontier LLMs: A Study of the New York Times v. OpenAI 2023 Lawsuit.
• Thomas Baumann. Universal Jailbreak Backdoors in Large Language Model Alignment.
• Kazuki Egashira. Exploring White-Box Model Access for Membership Inference.