相关论文: Constitutional Classifiers++: Efficient Production…

We propose an efficient scheme for verifying quantum computations in the `high complexity' regime i.e. beyond the remit of classical computers. Previously proposed schemes remarkably provide confidence against arbitrarily malicious…

In this paper we propose a novel defense approach against end-to-end adversarial attacks developed to fool advanced speech-to-text systems such as DeepSpeech and Lingvo. Unlike conventional defense approaches, the proposed approach does not…

Large Language Models (LLMs) have demonstrated remarkable performance across diverse tasks. Nevertheless, they still pose notable safety risks due to potential misuse for malicious purposes. Jailbreaking, which seeks to induce models to…

While significant attention has been dedicated to exploiting weaknesses in LLMs through jailbreaking attacks, there remains a paucity of effort in defending against these attacks. We point out a pivotal factor contributing to the success of…

Detecting jailbreak behaviour in large language models remains challenging, particularly when strongly aligned models produce harmful outputs only rarely. In this work, we present an empirical study of output based jailbreak detection under…

In this study, we introduce RePD, an innovative attack Retrieval-based Prompt Decomposition framework designed to mitigate the risk of jailbreak attacks on large language models (LLMs). Despite rigorous pretraining and finetuning focused on…

Large language models (LLMs) have achieved remarkable capabilities but remain vulnerable to adversarial prompts known as jailbreaks, which can bypass safety alignment and elicit harmful outputs. Despite growing efforts in LLM safety…

Automatic adversarial prompt generation provides remarkable success in jailbreaking safely-aligned large language models (LLMs). Existing gradient-based attacks, while demonstrating outstanding performance in jailbreaking white-box LLMs,…

While defenses against single-turn jailbreak attacks on Large Language Models (LLMs) have improved significantly, multi-turn jailbreaks remain a persistent vulnerability, often achieving success rates exceeding 70% against models optimized…

Compartmentalization is good security-engineering practice. By breaking a large software system into mutually distrustful components that run with minimal privileges, restricting their interactions to conform to well-defined interfaces, we…

While multimodal large language models (MLLMs) have achieved remarkable success in recent advancements, their susceptibility to jailbreak attacks has come to light. In such attacks, adversaries exploit carefully crafted prompts to coerce…

Recent research has demonstrated that state-of-the-art LLMs and defenses remain susceptible to multi-turn jailbreak attacks. These attacks require only closed-box model access and are often easy to perform manually, posing a significant…

Recently, Anil et al. (2024) show that many-shot (up to hundreds of) demonstrations can jailbreak state-of-the-art LLMs by exploiting their long-context capability. Nevertheless, is it possible to use few-shot demonstrations to efficiently…

Existing backdoor defense methods are only effective for limited trigger types. To defend different trigger types at once, we start from the class-irrelevant nature of the poisoning process and propose a novel weakly supervised backdoor…

As large language models (LLMs) are increasingly deployed in critical applications, ensuring their robustness and safety alignment remains a major challenge. Despite the overall success of alignment techniques such as reinforcement learning…

Text classifiers are vulnerable to adversarial examples -- correctly-classified examples that are deliberately transformed to be misclassified while satisfying acceptability constraints. The conventional approach to finding adversarial…

Recent work has shown that state-of-the-art classifiers are quite brittle, in the sense that a small adversarial change of an originally with high confidence correctly classified input leads to a wrong classification again with high…

Large Language Models face security threats from jailbreak attacks. Existing research has predominantly focused on prompt-level attacks while largely ignoring the underexplored attack surface of user-controlled response prefilling. This…

Large language models have seen widespread adoption, yet they remain vulnerable to multi-turn jailbreak attacks, threatening their safe deployment. This has led to the task of training automated multi-turn attackers to probe model safety…

Large Language Models (LLMs) are known to be vulnerable to jailbreaking attacks, wherein adversaries exploit carefully engineered prompts to induce harmful or unethical responses. Such threats have raised critical concerns about the safety…