评分 6.7 · 来源：cs.CL updates on arXiv.org · 发布于 2026-04-08
评分依据：LLM Agent红队测试新方法
arXiv:2604.05549v1 Announce Type: new Abstract: With the widespread application of LLM-based agents across various domains, their complexity has introduced new security threats. Existing red-team methods mostly rely on modifying user prompts, which lack adaptability to new data and may impact the agent’s performance. To address the challenge, this paper proposes the JailAgent framework, which completely avoids modifying the user prompt. Specifically, it implicitly manipulates the agent’s reasoning trajectory and memory retrieval with three key stages: Trigger Extraction, Reasoning Hijacking, a