Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 30 papers Another 265 papers do not have category tags yet and are not included in filtered results.
“
About this collection Collection in progress We are building a large AI safety paper collection to help researchers find work related to their interests. Papers are collected automatically andfiltered only for relevance to AI safety , without judging quality, novelty or reading value. The default “Annotated first” shows papers with tags or published paper analyses first. Each group is ordered by original publication date, newest first, with undated papers last. “Newest first” uses publication dates only. This order is not a quality ranking.
One-line summaries, tags and paper analyses are prepared by AI from the papers. Language models may have systematic biases and produce hallucinations.We plan to carry out human annotation and review to reduce errors in this content.
提出 PackHallu，经规则文件提示注入诱导编码 Agent 换用攻击者指定包
Published Oct 7
Evaluated models Qwen2.5-Coder-7B , Qwen3-Coder-30B , Devstral-Small-2-24B and others19 Abstract PackHallu 诱导规则文件换包。PackHallu 研究一种针对编程智能体规则文件的提示注入：让智能体在本应使用合法依赖时，改导入攻击者指定并控制的包。
提出 ASPIRE 行为图搜索，开放式发现 Agent 提示注入漏洞
Published Oct 7
Evaluated models Gemini-3.7-Flash , DeepSeek-V4-Flash , Gemini-2.5-flash Abstract ASPIRE 把提示注入红队从载荷优化改成行为图引导的开放式漏洞发现。ASPIRE 是给搭建方用的智能体提示注入红队引擎，目标是开放式画出可利用的行为路径，而不是在固定任务上打磨一条载荷。
提出 CORSA，按任务簇优化技能注入的检索与执行
Published Oct 6
Evaluated models GPT-5.4 , DeepSeek-V4-Pro , Qwen3.8-27B and others5 提出 APEX，用跨技能进度记录夹带虚假授权劫持 Agent
Published Oct 1
Evaluated models Claude Sonnet 5 , DeepSeek V4 Flash , GPT-5.4 and others6 提出 TrustProbe，挖掘技能智能体中不可信 skill 到特权操作的信任链漏洞
Published Sep 30
Evaluated models OpenClaw , OpenCode , Hermes Agent and others11 提出 AdaLCPI，把恶意指令拆成碎片嵌入工具返回并诱导 Agent 越权执行
Published Sep 29
Evaluated models GPT-4.1 , GPT-5.1 , GPT-5.6-Luna and others7 提出跨独立 Agent 的记忆跳跃攻击，经共享工件在助手间自传播
Published Sep 29
Evaluated models GPT-5.6 Luna , Kimi-K2.6 , GPT-OSS-120B and others6 Abstract 论文证实持久工件与内存可形成跨智能体传播链，揭示了将共享工件纳入安全防线的重要意义。该论文系统研究并量化了通过共享持久工件在具有独立私有内存的 LLM 智能体之间发生自传播的“工件介导传播”攻击风险。
用跨目标课程强化学习训练提示注入红队，攻破前沿智能体
Published Sep 27
Evaluated models GPT-4o-mini , GPT-5-nano , GPT-5.6-Luna and others12 Abstract 提出跨目标模型课程强化学习，攻破前沿模型冷启动瓶颈并实现跨模型迁移。该研究针对强化学习自动化提示注入红队在攻击前沿大模型时的冷启动问题，提出了一种基于目标模型鲁棒性递进的课程强化学习方法。
测量间接提示注入对类型化概率决策的动作劫持
Published Sep 24
Evaluated models jev-1.13.0 Abstract 封闭动作集改变提示注入的表现，但没有去掉概率被不可信内容移动的风险。作者在 jev-1.13.0 上研究 decision hijacking：不可信内容能否在用户任务和可选动作都不变时，把类型化决策推向攻击者目标。
提出两阶段框架 A2M，优化 MCP 工具元数据与返回载荷以劫持 Agent
Published Sep 23
Evaluated models GLM-4.6 , Qwen3-Max , DeepSeek-V3.1 and others5 提出 Repeat-After-Me 自适应视觉提示注入，诱导 VLM 智能体执行指定任务
Published Sep 4
Evaluated models Claude-Opus-4.7 , GPT-5.5 , Gemini-3.1-Pro and others6 Abstract 论文提出了针对黑盒 VLM 的前缀诱导自适应视觉提示注入，在多款模型上诱发了工具调用与隐私泄露。
提出 Search Harness，把间接提示注入做成测试时搜索并自适应利用
Published Sep 4
Evaluated models Claude Opus 4.8 , GPT-5.5 提出 ECLIPSE 双通道提示注入，劫持长程 Agent 的动作轨迹
Published Aug 31
Evaluated models DeepSeek , GPT-4o , Claude 4.8 Opus and others7 提出 ContextLeak，用强化学习优化恶意工具描述，诱导 Agent 外泄运行时上下文
Published Aug 28
Evaluated models Qwen-3-8B , GPT-OSS-20B , Gemma-4-E4B-it and others11 提出 PIMiner，用可复用策略库自动搜索并迁移针对智能体的提示注入
Published Aug 6
Evaluated models Gemini-2.5-Pro , DeepSeek-V4-Pro , GPT-4o-mini and others10 提出 LoginTrap，以网页弹窗诱导 Web Agent 提交敏感信息
Published Aug 5
Evaluated models GPT-4o , Gemini 3 Flash , Claude Sonnet 4 and others7 Abstract 揭示网络智能体在黑盒网页下的登录诱导风险，利用模糊测试生成话术并跨骨干和架构验证了隐私泄露威胁。
提出权威链劫持与轨迹引导策略演化，劫持搜索 Agent 的证据链
Published Aug 5
Evaluated models Qwen3.5-9B , Gemma4-31B , DeepResearch and others7 Abstract 论文揭示了多步搜索智能体在受限中间人操纵下的脆弱性，提出 ACH 策略与 TGSE 演化方法并完成系统验证。
构建 OpenART 竞技场，以 EMHA 演化持久环境评测智能体安全
Published Aug 1
Evaluated models GPT-5.5 , Claude-Opus-4.8 , GLM-5.2 and others5 训练自博弈红队智能体 GPT-Red，规模化生成提示注入与越狱攻击
Published Jul 29
Evaluated models GPT-5.6 , GPT-5.5 , GPT-5.4 and others8 提出 AHA 自动科研循环，发现生产 Agent 的可复用漏洞概念
Published Jul 13
Evaluated models Minimax-M2.7 , Kimi-K2.6 , Deepseek-V4-Pro and others5 Abstract 论文提出基于可证伪科研循环的 AHA 框架，从生产级智能体中挖掘并沉淀漏洞概念图，提升攻击复用率并指导防御加固。AHA 是一项针对生产级 LLM 智能体自动化红队测试的研究，旨在解决传统方法仅保留具体攻击载荷而缺乏成因解释、导致跨版本复用困难且无法指导修复的问题。