Related papers: Directional Embedding Smoothing for Robust Vision …
Large visual language models (LVLMs) have demonstrated excellent instruction-following capabilities, yet remain vulnerable to stealthy backdoor attacks when finetuned using contaminated data. Existing backdoor defense techniques are usually…
This paper proposes a guaranteed defense method for large language models (LLMs) to safeguard against jailbreaking attacks. Drawing inspiration from the denoised-smoothing approach in the adversarial defense domain, we propose a novel…
Multimodal large language models (MLLMs) have achieved impressive performance across diverse tasks by jointly reasoning over textual and visual inputs. Despite their success, these models remain highly vulnerable to adversarial…
Language models (LMs) and their extension, vision-language models (VLMs), have achieved remarkable performance across various tasks. However, they still struggle with complex reasoning tasks that require multimodal or multilingual…
Vision-language models (VLMs) have revolutionized multimodal AI applications but introduce novel security vulnerabilities that remain largely unexplored. We present the first comprehensive study of steganographic prompt injection attacks…
Stable pre-training is essential for achieving better-performing language models. However, tracking pre-training stability by calculating gradient variance at every step is impractical due to the significant computational costs. We explore…
As Large Language Models (LLMs) are widely used, understanding them systematically is key to improving their safety and realizing their full potential. Although many models are aligned using techniques such as reinforcement learning from…
Large Vision-Language Models (LVLMs) demonstrate exceptional performance across multimodal tasks, yet remain vulnerable to jailbreak attacks that bypass built-in safety mechanisms to elicit restricted content generation. Existing black-box…
Large language model (LLM)-powered multi-agent systems (MAS) enable agents to communicate and share information, achieving strong performance on complex tasks. However, this communication also creates an attack surface where malicious…
Multi-modal Large Language Models (MLLMs) excel in vision-language tasks but remain vulnerable to visual adversarial perturbations that can induce hallucinations, manipulate responses, or bypass safety mechanisms. Existing methods seek to…
We study typographic prompt injection attacks on vision-language models (VLMs), where adversarial text is rendered as images to bypass safety mechanisms, posing a growing threat as VLMs serve as the perceptual backbone of autonomous agents,…
Large Language Models (LLMs) and Vision-Language Models (VLMs) are increasingly deployed in robotic environments but remain vulnerable to jailbreaking attacks that bypass safety mechanisms and drive unsafe or physically harmful behaviors in…
Large language models (LLMs) excel in various tasks but remain vulnerable to jailbreak attacks, where adversaries manipulate prompts to generate harmful outputs. Examining jailbreak prompts helps uncover the shortcomings of LLMs. However,…
Vision language models (VLMs) excel in multimodal understanding but are prone to adversarial attacks. Existing defenses often demand costly retraining or significant architecture changes. We introduce a lightweight defense using tensor…
The widespread practice of fine-tuning open-source Vision-Language Models (VLMs) raises a critical security concern: jailbreak vulnerabilities in base models may persist in downstream variants, enabling transferable attacks across…
Foundation models based on large language models (LLMs) have shown great success in handling various tasks and modalities. However, adapting these models for general-purpose audio-language tasks is challenging due to differences in acoustic…
Recent advancements in Large Vision-Language Models (VLMs) have underscored their superiority in various multimodal tasks. However, the adversarial robustness of VLMs has not been fully explored. Existing methods mainly assess robustness…
As vision-language models (VLMs) gain prominence, their multimodal interfaces also introduce new safety vulnerabilities, making the safety evaluation challenging and critical. Existing red-teaming efforts are either restricted to a narrow…
Large Vision-Language Models (LVLMs) rely on attention-based retrieval of safety instructions to maintain alignment during generation. Existing attacks typically optimize image perturbations to maximize harmful output likelihood, but suffer…
This study evaluates the effectiveness of Vision Language Models (VLMs) in representing and utilizing multimodal content for fact-checking. To be more specific, we investigate whether incorporating multimodal content improves performance…