23 citations · 35 across the 8 of their papers we have counts for
9 papers
Steal the Patch Size: Adversarially Manipulate Vision-Language Models
Kai Hu, Akash Bharadwaj, Weichen Yu +1
We present a black-box model-stealing attack that recovers private vision-tokenizer configurations of deployed vision-language models (VLMs), including the visual patch size and in…
Configurable Reward Model for Balanced Safety Alignment
Zhengping Jiang, Mehran Khodabandeh, Akash Bharadwaj +5
Aligning large language models (LLMs) to heterogeneous and rapidly evolving safety requirements remains a critical challenge. Existing instruction-tuned LLMs and standalone safety…
Jailbreak-Zero: A Path to Pareto Optimal Red Teaming for Large Language Models
Kai Hu, Abhinav Aggarwal, Mehran Khodabandeh +6
This paper introduces Jailbreak-Zero, a novel red teaming methodology that shifts the paradigm of Large Language Model (LLM) safety evaluation from a constrained example-based appr…
Federated Experiment Design under Distributed Differential Privacy
Wei-Ning Chen, Graham Cormode, Akash Bharadwaj +2
Experiment design has a rich history dating back over a century and has found many critical applications across various fields since then. The use and collection of users' data in…
When does the student surpass the teacher? Federated Semi-supervised Learning with Teacher-Student EMA
Jessica Zhao, Sayan Ghosh, Akash Bharadwaj +1
Semi-Supervised Learning (SSL) has received extensive attention in the domain of computer vision, leading to development of promising approaches such as FixMatch. In scenarios wher…
The communication cost of security and privacy in federated frequency estimation
Wei-Ning Chen, Ayfer Özgür, Graham Cormode +1
We consider the federated frequency estimation problem, where each user holds a private item from a size- domain and a server aims to estimate the empirical frequency (i.e…
Sample and Threshold Differential Privacy: Histograms and applications
Akash Bharadwaj, Graham Cormode
Federated analytics seeks to compute accurate statistics from data distributed across users' devices while providing a suitable privacy guarantee and being practically feasible to…
Opacus: User-Friendly Differential Privacy Library in PyTorch
Ashkan Yousefpour, Igor Shilov, Alexandre Sablayrolles +9
We introduce Opacus, a free, open-source PyTorch library for training deep learning models with differential privacy (hosted at opacus.ai). Opacus is designed for simplicity, flexi…
To Test Machine Comprehension, Start by Defining Comprehension
Jesse Dunietz, Gregory Burnham, Akash Bharadwaj +3
Many tasks aim to measure machine reading comprehension (MRC), often focusing on question types presumed to be difficult. Rarely, however, do task designers start by considering wh…