6 papers
VaultGemma: A Differentially Private Gemma Model
Amer Sinha, Thomas Mesnard, Ryan McKenna +18
We introduce VaultGemma 1B, a 1 billion parameter model within the Gemma family, fully trained with differential privacy. Pretrained on the identical data mixture used for the Gemm…
Balls-and-Bins Sampling for DP-SGD
Lynn Chua, Badih Ghazi, Charlie Harrison +6
We introduce the Balls-and-Bins sampling for differentially private (DP) optimization methods such as DP-SGD. While it has been common practice to use some form of shuffling in DP-…
Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models
Lynn Chua, Badih Ghazi, Yangsibo Huang +6
Large language models (LLMs) are typically multilingual due to pretraining on diverse multilingual corpora. But can these models relate corresponding concepts across languages, i.e…
Scaling Laws for Differentially Private Language Models
Ryan McKenna, Yangsibo Huang, Amer Sinha +9
Scaling laws have emerged as important components of large language model (LLM) training as they can predict performance gains through scale, and provide guidance on important hype…
Scalable DP-SGD: Shuffling vs. Poisson Subsampling
Lynn Chua, Badih Ghazi, Pritish Kamath +4
We provide new lower bounds on the privacy guarantee of the multi-epoch Adaptive Batch Linear Queries (ABLQ) mechanism with shuffled batch sampling, demonstrating substantial gaps…
Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy
Yangsibo Huang, Daogao Liu, Lynn Chua +7
Machine unlearning algorithms, designed for selective removal of training data from models, have emerged as a promising approach to growing privacy concerns. In this work, we expos…