多流扰动攻击:通过并发任务干扰破坏思考LLM的安全对齐
Multi-Stream Perturbation Attack: Breaking Safety Alignment of Thinking LLMs Through Concurrent Task Interference
AI总结:
多流扰动攻击通过并发任务干扰破坏思考LLM的安全对齐,实现高攻击成功率和推理过程崩溃
摘要:
大型语言模型(LLM)中思考模式的广泛应用显著增强了复杂任务处理能力,同时引入了新的安全风险。当受到劫持攻击时,逐步推理过程可能导致模型生成更详细有害内容。我们观察到,当处理交错的多个任务时,思考模式表现出独特的脆弱性。基于这一观察,我们提出了多流扰动攻击,通过在单个提示中交错多个任务流来生成叠加干扰。我们设计了三种扰动策略:多流交错、反向扰动和形状变换,分别通过并发任务交错、字符反转和格式约束来破坏推理过程。在JailbreakBench、AdvBench和HarmBench数据集上,我们的方法在主流模型包括Qwen3系列、DeepSeek、Qwen3-Max和Gemini 2.5 Flash上实现了攻击成功率超过大多数方法。实验显示,思考崩溃率和响应重复率分别达到17%和60%,表明多流扰动不仅绕过了安全机制,还导致推理过程崩溃或重复输出。
英文摘要:
The widespread adoption of thinking mode in large language models (LLMs) has significantly enhanced complex task processing capabilities while introducing new security risks. When subjected to jailbreak attacks, the step-by-step reasoning process may cause models to generate more detailed harmful content. We observe that thinking mode exhibits unique vulnerabilities when processing interleaved multiple tasks. Based on this observation, we propose multi-stream perturbation attack, which generates superimposed interference by interweaving multiple task streams within a single prompt. We design three perturbation strategies: multi-stream interleaving, inversion perturbation, and shape transformation, which disrupt the thinking process through concurrent task interleaving, character reversal, and format constraints respectively. On JailbreakBench, AdvBench, and HarmBench datasets, our method achieves attack success rates exceeding most methods across mainstream models including Qwen3 series, DeepSeek, Qwen3-Max, and Gemini 2.5 Flash. Experiments show thinking collapse rates and response repetition rates reach up to 17% and 60% respectively, indicating multi-stream perturbation not only bypasses safety mechanisms but also causes thinking process collapse or repetitive outputs.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。