Youssef Allouah

I am a postdoctoral researcher at Stanford University. I received my PhD in computer science from EPFL and graduated in mathematics and computer science from Ecole Polytechnique. I also interned at Amazon.

I study the foundations of trustworthy machine learning, developing methods with provable guarantees and understanding fundamental limits. My work spans data privacy, robustness to malicious participants, data deletion and machine unlearning, and reliable model evaluation.

Youssef Allouah

Recent Work

See the publications page for the complete list.

Fast but False Progress on Benchmarks with Richer Feedback

Y. Allouah, J. Duchi, S. Koyejo.

websitecode

How Reusable Are Benchmarks with Richer Feedback?

Y. Allouah, J. Duchi.

papercode

The Distillation Game: Adaptive Evaluations & Efficient Defenses

Y. Allouah*, M. Haghifam*, S. Koyejo, R. Shokri. (* equal contribution)

papercode

What to Forget in Unlearning? Forget Set Curation for Language Models

A. Jha, A. Khatua, Y. Allouah, S. Koyejo.

Conference on Neural Information Processing Systems (NeurIPS), 2026

papercode

Distributional Machine Unlearning via Selective Data Removal

Y. Allouah, R. Guerraoui, S. Koyejo.

International Conference on Learning Representations (ICLR), 2026

ICML Workshop on Machine Unlearning for Generative AI (MUGen), 2025 (Oral)

papercode