相关论文: RAJ-PGA: Reasoning-Activated Jailbreak and Princip…

Large Language Models (LLMs) have increasingly become pivotal in content generation with notable societal impact. These models hold the potential to generate content that could be deemed harmful.Efforts to mitigate this risk include…

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…

Large Reasoning Models (LRMs) have recently extended their powerful reasoning capabilities to safety checks-using chain-of-thought reasoning to decide whether a request should be answered. While this new approach offers a promising route…

While large language models (LLMs) have achieved remarkable success in providing trustworthy responses for knowledge-intensive tasks, they still face critical limitations such as hallucinations and outdated knowledge. To address these…

As large language models (LLMs) are increasingly deployed, ensuring their safe use is paramount. Jailbreaking, adversarial prompts that bypass model alignment to trigger harmful outputs, present significant risks, with existing studies…

Large Language Models (LLMs) are increasingly susceptible to jailbreak attacks, which are adversarial prompts that bypass alignment constraints and induce unauthorized or harmful behaviors. These vulnerabilities undermine the safety,…

Aligning large language models (LLMs) with human values, particularly when facing complex and stealthy jailbreak attacks, presents a formidable challenge. Unfortunately, existing methods often overlook this intrinsic nature of jailbreaks,…

Large Reasoning Models (LRMs) have advanced automated multi-step reasoning, but their ability to generate complex Chain-of-Thought (CoT) trajectories introduces severe privacy risks, as sensitive information may be deeply embedded…

Retrieval Augmented Generation (RAG) has emerged as the de facto industry standard for user-facing NLP applications, offering the ability to integrate data without re-training or fine-tuning Large Language Models (LLMs). This capability…

Retrieval-Augmented Generation (RAG) systems have emerged as a promising solution to mitigate LLM hallucinations and enhance their performance in knowledge-intensive domains. However, these systems are vulnerable to adversarial poisoning…

We study a new vulnerability in commercial-scale safety-aligned large language models (LLMs): their refusal to generate harmful responses can be broken by flipping only a few bits in model parameters. Our attack jailbreaks billion-parameter…

Large language models (LLMs) have achieved remarkable success due to their exceptional generative capabilities. Despite their success, they also have inherent limitations such as a lack of up-to-date knowledge and hallucination.…

Large Reasoning Models (LRMs) achieve remarkable success through explicit thinking steps, yet the thinking steps introduce a novel risk by potentially amplifying unsafe behaviors. Despite this vulnerability, conventional defense mechanisms…

Aligned large language models (LLMs) remain vulnerable to jailbreak attacks. Recent mechanistic studies have identified latent features and representation shifts associated with jailbreak success, but they leave a more fundamental question…

In the realm of black-box jailbreak attacks on large language models (LLMs), the feasibility of constructing a narrow safety proxy, a lightweight model designed to predict the attack success rate (ASR) of adversarial prompts, remains…

Currently, open-sourced large language models (OSLLMs) have demonstrated remarkable generative performance. However, as their structure and weights are made public, they are exposed to jailbreak attacks even after alignment. Existing…

Large Language Models (LLMs) have transformed task automation and content generation across various domains while incorporating safety filters to prevent misuse. We introduce a novel jailbreaking framework that employs distributed prompt…

Emerging large reasoning models (LRMs), such as DeepSeek-R1 models, leverage long chain-of-thought (CoT) reasoning to generate structured intermediate steps, enhancing their reasoning capabilities. However, long CoT does not inherently…

Large language models (LLMs) have notably progressed in multi-step and long-chain reasoning. However, extending their reasoning capabilities to encompass deep interactions with search remains a non-trivial challenge, as models often fail to…

Security alignment enables the Large Language Model (LLM) to gain the protection against malicious queries, but various jailbreak attack methods reveal the vulnerability of this security mechanism. Previous studies have isolated LLM…