While reasoning large language models (LLMs) demonstrate remarkable performance across various tasks, they also contain notable security vulnerabilities. Recent research has uncovered a "thinking-stopped" vulnerability in DeepSeek-R1, where model-generated reasoning tokens can forcibly interrupt the inference process, resulting in empty responses that compromise LLM-integrated applications. However, existing methods triggering this vulnerability require complex mathematical word problems with long prompts--even exceeding 5,000 tokens. To reduce the token cost and formally define this vulnerability, we propose a novel prompt injection attack named "Reasoning Interruption Attack", based on adaptive token compression. We demonstrate that simple standalone arithmetic tasks can effectively trigger this vulnerability, and the prompts based on such tasks exhibit simpler logical structures than mathematical word problems. We develop a systematic approach to efficiently collect attack prompts and an adaptive token compression framework that utilizes LLMs to automatically compress these prompts. Experiments show our compression framework significantly reduces prompt length while maintaining effective attack capabilities. We further investigate the attack's performance via output prefix and analyze the underlying causes of the vulnerability, providing valuable insights for improving security in reasoning LLMs.

本研究针对大型语言模型（LLM）存在的"思考停止"漏洞进行了深入探讨，提出了一种新颖的攻击方法——"推理中断攻击"，利用自适应令牌压缩简化了攻击的提示词结构。实验表明，该方法能有效减少提示词长度，同时保持攻击效果，对提升LLM的安全性具有重要的启示意义。

高效令牌注入攻击：通过自适应令牌压缩促使大型语言模型停止推理