47 citations · 272 across the 43 of their papers we have counts for
7 papers · 1 filter
Information Theoretic Adversarial Training of Large Language Models
Yiwei Zhang, Jeremiah Birrell, Reza Ebrahimi +3
Large language models (LLMs) remain vulnerable to adversarial prompting despite advances in alignment and safety, often exhibiting harmful behaviors under novel attack strategies.…
Learning to Forget with Information Divergence Reweighted Objectives for Noisy Labels
Jeremiah Birrell, Reza Ebrahimi
We introduce ANTIDOTE, a new class of objectives for learning under noisy labels which are defined in terms of a relaxation over an information-divergence neighborhood. Using conve…
An Interactive Framework for Implementing Privacy-Preserving Federated Learning: Experiments on Large Language Models
Kasra Ahmadi, Rouzbeh Behnia, Reza Ebrahimi +4
Federated learning (FL) enhances privacy by keeping user data on local devices. However, emerging attacks have demonstrated that the updates shared by users during training can rev…
Differentially Private Stochastic Gradient Descent with Fixed-Size Minibatches: Tighter RDP Guarantees with or without Replacement
Jeremiah Birrell, Reza Ebrahimi, Rouzbeh Behnia +1
Differentially private stochastic gradient descent (DP-SGD) has been instrumental in privately training deep learning models by providing a framework to control and track the priva…
Optimal Transport Regularized Divergences: Application to Adversarial Robustness
Jeremiah Birrell, Reza Ebrahimi
We introduce a new class of optimal-transport-regularized divergences, , constructed via an infimal convolution between an information divergence, , and an optimal-transpor…
Structure-preserving GANs
Jeremiah Birrell, Markos A. Katsoulakis, Luc Rey-Bellet +1
Generative adversarial networks (GANs), a class of distribution-learning methods based on a two-player game between a generator and a discriminator, can generally be formulated as…