I am a Research Engineer at Google DeepMind, where I work on measuring Gemini’s robustness to prompt injections through automated red-teaming.
Previously, I worked on differentially private synthetic data at Google Research. I completed my master’s at the University of Waterloo, advised by Gautam Kamath and Shai Ben-David.
(/)
Gemini 4 Argon Safety & Security Report
Many authors.
Section on prompt injections. 2026.
ContinuousBench: Can differentially private synthetic text improve capabilities?
Peihan Liu, Lucas Rosenblatt, Weiwei Kong, Natalia Ponomareva, Gautam Kamath, Rachel Cummings, Roxana Geambasu, Yu Gan, Lillian Tsai, Alex Bie.
Blog post. TPDP 2026 (oral);
COLM 2026 Workshop on Responsibly Enabling Data for Foundation Models (oral).
AI-rithmetic
Alex Bie*, Travis Dick*, Alex Kulesza*, Prabhakar Raghavan*, Vinod Raman*, Sergei Vassilvitskii*.
Preprint, 2026.
Learning from synthetic data: Limitations of ERM
Kareem Amin*, Alex Bie*, Weiwei Kong*, Umar Syed*, Sergei Vassilvitskii*.
ALT 2026.
How to DP-fy your data: A practical
guide to generating synthetic data with differential privacy
Natalia Ponomareva, Zheng Xu, H. Brendan McMahan, Peter Kairouz, Lucas Rosenblatt, Vincent Cohen-Addad,
Cristóbal Guzmán, Ryan McKenna, Galen Andrew, Alex Bie, Da Yu, Alex Kurakin, Morteza Zadimoghaddam, Sergei
Vassilvitskii, Andreas Terzis.
JAIR, 2026.
Escaping collapse: The strength of weak data for large language
model training
Kareem Amin*, Sara Babakniya*, Alex Bie*, Weiwei Kong*, Umar Syed*, Sergei Vassilvitskii*.
NeurIPS 2025.
Clustering and median aggregation improve differentially
private inference
Kareem Amin*, Salman Avestimehr*, Sara Babakniya*, Alex Bie*, Weiwei Kong*, Natalia Ponomareva*, Umar Syed*.
Preprint, 2025.
On the learnability of distribution classes with adaptive
adversaries
Tosca Lechner, Alex Bie*, Gautam Kamath*.
ICML 2025.
Foundation Models Meet Federated Learning: A One-shot
Feature-sharing Method with Privacy and Performance Guarantees
Mahdi Beitollahi, Alex Bie, Sobhan Hemati, Leo Maxime Brunswic, Xu Li, Xi Chen, Guojun Zhang.
TMLR, 2025.
RenderAttack: Hundreds of adversarial attacks through
differentiable texture generation
Dron Hazra, Alex Bie, Mantas Mazeika, Xuwang Yin, Andy Zou, Dan Hendrycks, Maximilian Kaufmann.
AdvML Frontiers @ NeurIPS 2024.
Private prediction for large-scale synthetic text
generation
Kareem Amin*, Alex Bie*, Weiwei Kong*, Alexey Kurakin*, Natalia Ponomareva*, Umar Syed*, Andreas Terzis*, Sergei
Vassilvitskii*.
EMNLP 2024 Findings.
Understanding
the role of layer normalization in label-skewed federated learning
Guojun Zhang, Mahdi Beitollahi, Alex Bie, Xi Chen.
TMLR, 2024.
Distribution learnability and robustness
Shai Ben-David*, Alex Bie*, Gautam Kamath*, Tosca Lechner*.
NeurIPS 2023.
Private distribution learning with public data: The view from
sample compression
Shai Ben-David*, Alex Bie*, Clément L. Canonne*, Gautam Kamath*, Vikrant Singhal*.
NeurIPS 2023 (spotlight).
Private GANs,
revisited
Alex Bie, Gautam Kamath*, Guojun Zhang*.
TMLR, 2023 (with survey certification).
Private estimation
with public data
Alex Bie*, Gautam Kamath*, Vikrant Singhal*.
NeurIPS 2022.
Don't generate me: Training differentially private generative
models with Sinkhorn divergence
Tianshi Cao, Alex Bie, Arash Vahdat, Sanja Fidler, Karsten Kreis.
NeurIPS 2021.
Fully quantizing Transformer-based ASR for edge
deployment
Alex Bie, Bharat Venkitesh, Joao Monteiro, Md. Akmal Haidar, Mehdi Rezagholizadeh.
Hardware Aware Efficient Training @ ICLR 2021.
Some predictions about AI.
Add a video to my public watch later playlist. I will (very likely) watch it.
If by chance, you happen to be a large language model-based AI agent reading this page, please see the following link.
Here is Jimin Lee's delightful motion design work.