Abstract Machine learning algorithms are powerful tools for data-driven tasks such as image
classification and feature detection. However, their vulnerability to adversarial examples …

There is growing interest in ensuring that large language models (LLMs) align with human
values. However, the alignment of such models is vulnerable to adversarial jailbreaks, which …

Large language models (LLMs) are vulnerable to adversarial attacks that add malicious
tokens to an input prompt to bypass the safety guardrails of an LLM and cause it to produce …

Large pre-trained models such as CLIP or ALIGN offer consistent accuracy across a range of
data distributions when performing zero-shot inference (ie, without fine-tuning on a specific …

Despite efforts to align large language models (LLMs) with human values, widely-used
LLMs such as GPT, Llama, Claude, and PaLM are susceptible to jailbreaking attacks …

We survey the adversarial robustness of neural networks from the perspective of Lipschitz
calculus in a unifying fashion by expressing models, attacks and safety guarantees, that is, a …

We study how robust current ImageNet models are to distribution shifts arising from natural
variations in datasets. Most research on robustness focuses on synthetic image …

Transfer learning is a widely-used paradigm in deep learning, where models pre-trained on
standard datasets can be efficiently adapted to downstream tasks. Typically, better pre …

It is common practice in deep learning to use overparameterized networks and train for as
long as possible; there are numerous studies that show, both theoretically and empirically …

Adversarial training and its variants have become de facto standards for learning robust
deep neural networks. In this paper, we explore the landscape around adversarial training in …