Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 12 papers Another 232 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 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。
评测多轮对话中 GradSafe 越狱检测的可分性与脆弱性
Published Sep 29
Evaluated models Llama 3.1 8B Instruct , Qwen2.5-7B-Instruct Abstract CWS 显示不安全梯度在真实多轮流量中变弱，且随攻击族和模型而变。作者评估梯度越狱检测从单条提示词迁到多轮对话后还剩多少可分性，方法是用 Context Window Scanner 把 GradSafe 原封不动地套到用户轮窗口上。
提出技能级联攻击，把有害目标拆散到多个技能里绕过单技能扫描
Published Sep 25
Evaluated models GPT-5.4 , GPT-5.4 Mini , Claude Opus 4.6 and others11 Abstract 论文探讨了多技能智能体系统面临的技能级联攻击风险，提出自动化红队框架与基准集，指出跨技能交互存在安全防护盲区。
提出 CompoSkill，将逐个通过扫描的技能编排为攻击链
Published Aug 17
Evaluated models GPT-5.4 , Gemini-3.1-flash , DeepSeek-V4 and others4 Abstract 论文分析了智能体技能组合风险，指出单包安全认证状态不具备可组合性，需转向路径级防护。
提出最小范数攻击集成，用鲁棒性曲线评测图像分类器
Published Jul 22
Evaluated models C1–C20（CIFAR-10，Table 1） , I1–I10（ImageNet，Table 1） Abstract 预算可控的最小范数集成用 AOI 逼近攻击前沿，DOI 给出不依赖单一ε的防御排序。这篇工作把对抗鲁棒评测从单一 ε、固定集成，改成可按查询预算收紧的曲线估计。
提出 PaperGuard，用图文注入与梯度扰动抬高自动审稿分数并检测恶意指令
Published Jun 11
Evaluated models Claude-sonnet-4.5 , GPT-4o , GPT-4.1 and others11 Abstract PaperGuard 测到提示注入与白盒视觉抬分，并比较分块检索防御。PaperGuard 同时评测并检测多模态 AI 审稿中的评分操纵。
构建 MultiTurnPSB，评测四轮医疗越狱与分类器输入防御
Published May 30
Evaluated models GPT-4.1-mini , Claude Sonnet 4.5 Abstract 多轮攻击加剧医疗模型风险，分类器干预有效但受高误报率制约。MultiTurnPSB 针对患者端医疗大模型在持续交互中的安全风险，构建了包含三种攻击模态的四轮对抗评测基准。
Using ADVERSA to measure guardrail degradation and reviewer reliability under multi-turn attacks
Published Mar 10
Evaluated models Claude Opus 4.6 , Gemini 3.1 Pro , GPT-5.2 Abstract ADVERSA measures multi-turn guardrail changes using per-turn compliance trajectories and three-reviewer consensus.ADVERSA is an automated multi-turn red-teaming measurement framework that records guardrail behavior as per-turn compliance trajectories and simultaneously measures whether reviewers agree. The problem it addresses: single-turn binary pass/fail cannot show how safety properties change under sustained adversarial pressure, nor can it separate gradual refusal consolidation from stable scores. The authors argue that real-world adversaries rewrite and persist over multiple turns.
提出纯视觉越狱攻击 VJA，把有害编辑指令编码进图像
Published Feb 11
Evaluated models Nano Banana Pro , GPT Image 1.5 , Qwen-Image-Edit and others9 提出 MM-Plan 多模态越狱框架，自动规划多轮图像编辑以越狱视觉独占目标
Published Feb 5
Evaluated models Llama-3.2-11B , InternVL3-8B , Qwen3-VL-8B and others8 Abstract 作者称全局规划能攻破对单轮攻击更稳的前沿 MLLM。MM-Plan 是面向 Visual Exclusivity 的 black-box 多模态红队方法，并配有人工基准 VE-Safety。
提出 DREAM 框架，评测 Agent 的跨环境长链攻击脆弱性
Published Dec 22, 2025
Evaluated models Gemini-2.5-Flash-NT , o4-mini , Gemini-2.5-Flash and others12 Abstract DREAM 提出基于知识图谱与策略搜索的动态红队框架，分析了长链跨环境攻击对当前主流智能体的系统性威胁。
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