Large language models (LLMs) can often be made to behave in undesirable ways that they
are explicitly fine-tuned not to. For example, the LLM red-teaming literature has produced a …

We conduct experiments on the impact of increasing inference-time compute in reasoning
models (specifically OpenAI o1-preview and o1-mini) on their robustness to adversarial …

Adversarial training (AT) refers to integrating adversarial examples--inputs altered with
imperceptible perturbations that can significantly impact model predictions--into the training …

Transformer-based models are driving a significant revolution in the field of machine
learning at the moment. Among these innovations, vision transformers (ViTs) stand out for …

Multi-modal Large Language Models (MLLMs) excel in vision-language tasks but remain
vulnerable to visual adversarial perturbations that can induce hallucinations, manipulate …

This paper investigates the robustness of vision-language models against adversarial visual
perturbations and introduces a novel``double visual defense" to enhance this robustness …

Large pre-trained Vision-Language Models (VLMs) such as CLIP have demonstrated
excellent zero-shot generalizability across various downstream tasks. However, recent …

Vision Transformers (ViTs) achieve excellent performance in various tasks, but they are also
vulnerable to adversarial attacks. Building robust ViTs is highly dependent on dedicated …

Adversarial attacks can be extremely dangerous, particularly in scenarios where the
precision of facial expression identification is of utmost importance. Hiring adversarial …