Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
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.
实现 DEFER，以确定性检查优先拦截多智能体流水线的内容攻击
Published Oct 6
Evaluated models Qwen3-235B-A22B-Instruct-2507 , gpt-oss-120b , Llama-4-Scout-17B-16E-Instruct and others5 Abstract DEFER 把确定性检查放在 judge 之前。DEFER 用来测量多智能体授权流水线里，确定性规则在何处结束、LLM judge 从何处开始。
提出检测、结构化与 GSPO 流程，降低共识生成的提示注入错位
Published Oct 3
Evaluated models LLaMA-2 7B , LLaMA-3 8B , GPT 3.5 Turbo and others14 Abstract 三章依次做个体偏好预测、政党知识图谱审议，以及共识生成对提示注入的脆弱性与一条防御流程。这是一篇把 LLM 数字孪生放进增强民主的博士论文，分个体表示、政党聚合和共识生成的提示注入三章。
提出 memetic trojans，借助社交传染在 Agent 网络中传播对抗载荷
Published Oct 1
Evaluated models gpt-oss-120B , deepseek-v4.1-flash , qwen-32B and others5 Abstract memetic trojans 用内生社会传染携带载荷。Memetic trojans 是一类网络介导攻击：对抗载荷搭在智能体本来就会转发和点赞的内容上扩散，转发者不必被攻陷，也不必执行“请继续复制”的指令。
提出 A2A-TIBA，经 A2A 对等任务间接注入并植入常驻回调
Published Sep 30
Evaluated models MiniMax-M3 , deepseek-v4-pro , glm-5.3 and others8 Abstract A2A-TIBA 一次植入后绕过智能体，GDA 用四类结果定位防线，ELA-ITL 把信封标注列为第三层。A2A-TIBA 把间接提示注入和绕过放进 A2A：gray-box 攻击者已能投递任务，诱导目标部署常驻回调程序后，下令和外传不再经过智能体。
提出 Skill-as-API，隔离多智能体协作中的 skill 正文与系统提示词
Published Sep 1 Abstract 协议把 skill 正文留在所有者侧，四层限制发现内容、权限期限和注入拼接。Skill-as-API 是面向智能体软件工程的协调协议，用来在跨组织调用时把 skill 的代码和系统提示词留在所有者进程。作者针对三条威胁：经通道或命名空间抽取专有方法，用注入让被调用模型输出系统提示词，以及协作结束后权限继续保留。
提出 Proof Gateway，用零知识谓词证明实现 Agent 间数据最小化
Published Aug 31 Abstract 作者将 source integrity 而不是证明开销视为绑定约束。Proof Gateway 是一套已运行的智能体间数据最小化机制：用治理谓词的零知识证明代替原始值和自然语言自述。
提出 SciTrace，为科学发现智能体加入跨阶段安全推理与工具轨迹核验
Published Jun 6
Evaluated models Llama-3.1-70B-Instruct , Qwen2.5-72B-Instruct , DeepSeek-V3 and others4 Abstract SciTrace 把安全推理织入科学智能体各阶段，在 SciSafetyBench 上提高安全指标并保持输出质量。SciTrace 是加在科学发现智能体流水线上的安全框架，不是再在各阶段出口放一个独立过滤器。
提出谱系图治理层抑制多智能体协作的错误级联
Published Mar 4 Abstract 单一种子可沿固定协作图固化为错误共识。作者把 LLM-MAS 中“小错变成集体错误共识”做成图上的可测传播，并用只改消息、不改拓扑的插件压低最终感染。
Defenses arXiv 2505.23559 · May 29, 2025 提出 SafeScientist 四层监控，阻止科研智能体执行高风险或不伦理任务
Published May 29, 2025
Evaluated models GPT-4o , GPT-4.1 , GPT-3.5-turbo and others5 Abstract 四层监控加科研安全基准。SafeScientist 是带四层监控的科研智能体框架，并用 SciSafetyBench 评测科学场景中的任务安全。
You have reached the end