Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 17 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 CredLeak-Bench,评测 Agent 面对钓鱼的凭据泄露与恢复
Published Oct 6
Evaluated models Claude Sonnet 5 , Claude Opus 4.8 , Gemini 3.6 Flash and others7 构建 AGENTDOXX 基准,评测联网搜索 Agent 对匿名访谈的再识别
Published Oct 5
Evaluated models GPT-5.6 Terra , GPT-5 Mini , GPT-4.1 Mini and others10 Abstract 论文提出了已知身份的重识别基准 AGENTDOXX,揭示搜索与记忆双重风险及现有防御在检索轨迹中的泄露漏洞。
构建网页浏览 Agent 的多语言多模态压力测试基准 HyperBrowseComp
Published Oct 3
Evaluated models Gemini 3.1 Pro Preview , Gemini 3.7 Flash , GPT-5.6 Terra and others5 构建 AGENTTELL 基准,测量浏览器 Agent 的行为侧信道隐私泄露
Published Sep 27
Evaluated models Claude Sonnet 5 , Gemini 3.7 Flash , GPT-5.6 Luna and others6 Abstract 论文评估了浏览器智能体的行为侧信道泄露风险,并基于近万次会话提供了评测数据。
提出 CAVEAT 基准,评测激励失配市场中计算机使用 Agent 的购买决策退化
Published Sep 23
Evaluated models GPT-5.6-Sol , GPT-5.6-Terra , GPT-5.6-Luna and others14 Abstract 提出 CAVEAT 基准揭示智能体在商业偏好下的决策退化,诊断出三类失效并给出缓解方案。
评测安全 Agent 在对抗性任务污染下的解题成功与开销
Published Sep 17
Evaluated models Gemma-4 31B , DeepSeek-V4 Flash , GPT-OSS 120B and others6 Abstract AgentLSD 用配对 web CTF 表明,欺骗性环境证据会改变安全智能体的解题率与工作量。AgentLSD 是一个对照框架,用来测量安全智能体在任务本身不变时,对对抗性任务污染的反应。
用 MisKnow-Agent 评测深度研究智能体是否采纳误导文档结论
Published Jul 23
Evaluated models DeerFlow , WebThinker , Gemini Deep Research and others6 Abstract 论文通过 MisKnow-Agent 框架评估发现长程研究智能体易采纳误导性知识,且现有防御难以完全消除该现象。
构建 Universal Verifier 核验计算机使用 Agent 轨迹
Published Apr 5
Evaluated models GPT-5.2 , GPT-4o , o4-mini and others8 Abstract 四条设计原则叠成 Universal Verifier,作者称其与人类一致程度相当且假阳性近零。Universal Verifier 给计算机使用轨迹打过程分和结果分,并定位失败。CUAVerifierBench 用人工标签衡量这种验证器。
提出 MalURLBench 评测 Agent 是否接受结构伪装的恶意 URL
Published Jan 26
Evaluated models GPT-3.5-Turbo , GPT-4o-mini , GPT-4o and others13 构建 TRAP 基准,评测网页 Agent 被界面提示注入劫持任务的情况
Published Dec 29, 2025
Evaluated models GPT-5 , Claude Sonnet 3.7 , Gemini 2.5 Flash and others6 Abstract 论文提出了基于五维模块化攻击空间的 TRAP 基准,评估了六个模型在真实网页环境中的注入风险。研究定位 :针对网页自主智能体易受外部界面提示注入攻击的问题,构建了基于真实网站克隆的模块化评测基准 TRAP。
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 提出 WAInjectBench 评测网页 Agent 的提示注入检测
Published Oct 2, 2025
Evaluated models GPT-4o , Mistral-7B-Instruct-v0.1 , Llama-Prompt-Guard-2-86M and others8 Abstract WAInjectBench 比较网页智能体的提示注入检测:显式线索可被部分检出,隐蔽攻击大多漏检。WAInjectBench 是网页智能体提示注入检测的基准。作者要看现有检测器能否发现这类攻击:攻击者改网页,使智能体执行目标动作,而不是用户任务。作者称这是首个此类综合基准。
提出网页 Agent 提示注入基准 WASP 并端到端评测安全性
Published Apr 23, 2025
Evaluated models GPT-4o , GPT-4o-mini , OpenAI o1 and others5 Abstract 论文提出针对网络智能体的端到端注入评测基准 WASP,作者称智能体易偏离目标但难以完成复杂攻击。
提出 SafeArena 基准,评测网页 Agent 是否执行故意有害任务
Published Mar 7, 2025
Evaluated models GPT-4o , Claude-3.5-Sonnet , Qwen-2-VL-72B and others5 Abstract SafeArena 用配对任务和 ARIA 评测网页智能体的故意滥用。SafeArena 评测自主网页智能体是否会执行用户故意给出的恶意网页任务。 。
提出 BrowserART,评测拒答训练在浏览器 Agent 上的泛化
Published Oct 11, 2024
Evaluated models o1-preview , o1-mini , GPT-4o (gpt-4o-2024-08-06) and others8 构建评测 Web Agent 安全与策略遵守的基准 ST-WebAgentBench
Published Oct 9, 2024 提出 ARE 框架与图像文本攻击,评测多模态网页 Agent 的对抗鲁棒性
Published Jun 19, 2024
Evaluated models GPT-4o , GPT-4V , Claude-3-Opus and others4 Abstract 系统评估多模态网页智能体的脆弱性,指出推理时计算组件被攻破会加剧系统风险。
You have reached the end