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。
构建支付欺诈基准 ACB，评测花钱智能体的欺诈检测
Published Sep 27
Evaluated models gordonguard , pattern judge Abstract ACB 把支付欺诈与智能体操纵分开测量，作者强调观察权限、误报预算和微支付复核成本。ACB 是一套智能体支付欺诈基准与检测参考栈 ，目标是衡量钱是否被不当提取，而不仅是智能体是否接受了恶意指令。
提出 TACIT，从冻结内部状态检测智能体有害轨迹
Published Sep 27
Evaluated models Qwen3-4B , Llama-3.1-8B , Qwen3-4B-Instruct and others4 Abstract 提出基于内部表征的智能体轨迹安全检测方法 TACIT，揭示工具风险表征特性并在多项基准上提升检测表现与效率。
提出 ATF 与 NTR，过滤或恢复工具集以阻止 Agent 调用攻击工具
Published Sep 14
Evaluated models Gemma2-9B , Qwen2-7B , LLaMA3-8B and others7 提出 ActGuard 执行前动作审计，定位并遮盖间接提示注入的恶意片段
Published Sep 14
Evaluated models GPT-4o-mini , Gemini-2.5-Flash , DeepSeek-V4-Flash 提出 MTGuard，对照 MCP 调用声明与执行痕迹拦截不安全工具调用
Published Jul 28
Evaluated models DeepSeek-V4-Flash , DeepSeek-V4-Pro , GPT-5.6 Luna 提出进程外双通道对比分类，拦截计算机使用智能体的错误点击
Published Mar 16
Evaluated models multi-modal-embed-small Abstract 感知错误可被屏幕像素对调利用。作者把 CUA 的感知失败写成安全漏洞visual confused deputy ：智能体按看错的屏幕授权 click(x,y)，真实落点与它以为的元素不一致。
提出 TS-Guard 与 TS-Flow，在执行前检测并引导 Agent 修正有害工具调用
Published Jan 15
Evaluated models GPT-4o , TS-Guard , Qwen2.5-14B-Instruct Abstract TS-Bench 标步骤级风险，TS-Guard 用多任务 RL 做执行前判定，TS-Flow 用反馈代替直接中止。ToolSafe 研究 LLM 智能体在工具执行前的步骤级安全：基准 TS-Bench、护栏 TS-Guard，以及反馈驱动框架 TS-Flow。
Defenses arXiv 2511.20597 · Nov 26, 2025 提出 BrowseSafe，在工具输出边界检测并拦截浏览器 Agent 提示注入
Published Nov 26, 2025
Evaluated models Qwen3-30B-A3B-Instruct-2507 , Qwen3-8B , PromptGuard-2 (22M) and others10 You have reached the end