Related papers: PRISM: Programmatic Reasoning with Image Sequence …
We discover a novel and surprising phenomenon of unintentional misalignment in reasoning language models (RLMs), which we call self-jailbreaking. Specifically, after benign reasoning training on math or code domains, RLMs will use multiple…
Large Language Models (LLMs) are increasingly attracting attention in various applications. Nonetheless, there is a growing concern as some users attempt to exploit these models for malicious purposes, including the synthesis of controlled…
Large language models (LLMs) are susceptible to a type of attack known as jailbreaking, which misleads LLMs to output harmful contents. Although there are diverse jailbreak attack strategies, there is no unified understanding on why some…
We introduce \emph{self-jailbreaking}, a threat model in which an aligned LLM guides its own compromise. Unlike most jailbreak techniques, which often rely on handcrafted prompts or separate attacker models, self-jailbreaking requires no…
Vision-language models (VLMs) have become central to tasks such as visual question answering, image captioning, and text-to-image generation. However, their outputs are highly sensitive to prompt variations, which can reveal vulnerabilities…
Large Language Models (LLMs) are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety…
Large Language Model (LLM) alignment remains vulnerable to jailbreak attacks that elicit unsafe responses, motivating pre-model and post-model guards. Pre-model guards audit the safety of prompts before invoking target models. However,…
Despite the implementation of safety alignment strategies, large language models (LLMs) remain vulnerable to jailbreak attacks, which undermine these safety guardrails and pose significant security threats. Some defenses have been proposed…
Large Language Models (LLMs) have been equipped with safety mechanisms to prevent harmful outputs, but these guardrails can often be bypassed through "jailbreak" prompts. This paper introduces a novel graph-based approach to systematically…
Large Reasoning Models (LRMs) face a distinct safety vulnerability: their internal reasoning chains may generate harmful content even when the final output appears benign. To address this overlooked risk, we first propose a novel attack…
Large Language Models (LLMs) have demonstrated impressive capabilities in natural language tasks, but their safety and morality remain contentious due to their training on internet text corpora. To address these concerns, alignment…
The proliferation of Large Language Models (LLMs) has revolutionized natural language processing and significantly impacted code generation tasks, enhancing software development efficiency and productivity. Notably, LLMs like GPT-4 have…
Large language models (LLMs) have achieved impressive performance across natural language tasks and are increasingly deployed in real-world applications. Despite extensive safety alignment efforts, recent studies show that such alignment is…
The emergence of Multimodal Large Language Models (MLRMs) has enabled sophisticated visual reasoning capabilities by integrating reinforcement learning and Chain-of-Thought (CoT) supervision. However, while these enhanced reasoning…
Multimodal large language models (MLLMs) are widely used in vision-language reasoning tasks. However, their vulnerability to adversarial prompts remains a serious concern, as safety mechanisms often fail to prevent the generation of harmful…
Multimodal Large Language Models (MLLMs) extend text-only LLMs with visual reasoning, but also introduce new safety failure modes under visually grounded instructions. We study comic-template jailbreaks that embed harmful goals inside…
We introduce new jailbreak attacks on vision language models (VLMs), which use aligned LLMs and are resilient to text-only jailbreak attacks. Specifically, we develop cross-modality attacks on alignment where we pair adversarial images…
Large language models (LLMs) have demonstrated remarkable capabilities, yet they also introduce novel security challenges. For instance, prompt jailbreaking attacks involve adversaries crafting sophisticated prompts to elicit responses from…
In recent years, the rapid development of large language models (LLMs) has achieved remarkable performance across various tasks. However, research indicates that LLMs are vulnerable to jailbreak attacks, where adversaries can induce the…
Considerable research efforts have been devoted to ensuring that large language models (LLMs) align with human values and generate safe text. However, an excessive focus on sensitivity to certain topics can compromise the model's robustness…