Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 22 papers Another 255 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.
评估应用决策层的输入注入与隐私推断风险
Published Oct 4
Evaluated models Jev , NanoJev Abstract Jev 的类型化决策可被输入与训练操纵，也会泄露信息，并同时支持检测与滥用。Jev 是 TypeSafe 的 System One 模型，把应用 state 和固定问题变成 Choice、Noul 或 Score 的类型化决策与概率。作者用官方 API 和可改训练的 NanoJev，考察它作为决策层时的操纵、泄露与双重用途。
提出 Persona Guardrail，强制客服智能体守住功能边界
Published Oct 2
Evaluated models Qwen3-30B-A3B , Qwen3-VL-30B , Qwen3-8B and others5 Abstract Persona Guardrail 用功能边界做同步输入输出校验，并在 PAGE 与生产延迟预算下给出评测。Persona Guardrail 是客服向智能体的同步运行时防御：先写明功能边界，再决定放行或拦截，而不是只检测某一攻击类。
提出 imMRAG 攻击，从多模态 RAG 私有图像库提取内容
Published Oct 1
Evaluated models Lumina-mGPT , Gemini 2.5 Flash Image Preview , CLIP ViT-B/16 and others5 提出 DART，用去噪表征转移检测多轮智能体攻击并注入提醒
Published Sep 30
Evaluated models Qwen3-8B , Qwen3-14B , Qwen3-32B and others6 提出能力洗钱攻击，拆分任务向对齐模型套取能力
Published Sep 14
Evaluated models GPT-5.5 , Claude Opus 4.8 , Grok-4.3 and others7 Abstract 作者提出了能力洗钱攻击，发现未对齐编排器分解咨询对齐模型可恢复能力差距，指出需建立跨会话防御。
提出 CIG-MIA，用上下文信息增益推断 RAG 知识库成员身份
Published Sep 14
Evaluated models Llama-3.1-8B-Instruct , Mistral-7B-Instruct-v0.3 , Llama-2-13b-chat-hf and others6 Abstract CIG-MIA 用注入前后的答案似然差推断 RAG 库成员，灰盒可分性高于所列基线。CIG-MIA 是针对 RAG 外部知识库的成员推断攻击，同时覆盖灰盒和纯文本黑盒。
Proposes SHAQ, which replaces document embeddings with shadow query embeddings to defend against embedding inversion.
Published Sep 4
Evaluated models GTR-T5 encoder , DPR Abstract SHAQ replaces document embeddings with shadow query embeddings.SHAQ is a defense against EIA for cloud vector databases: instead of directly modifying document embeddings, it replaces them with shadow query embeddings.
提出 RedEvoAgent，演化攻击技能对黑盒智能体做自动红队
Published Aug 28
Evaluated models MiniMax-M2.5 , DeepSeek-V4-Flash , Qwen3.5-35B Abstract RedEvoAgent 用验证棘轮演化可读攻击技能，在两种产品级 harness 上多数设置高于固定工具与红队基线。RedEvoAgent 是面向产品级 black-box 智能体的自动红队方法，用一份可演化、人类可读的攻击技能编排多个越狱工具。 。
提出 LeakGauge，用行为后缀的 prefill 对数概率检测上下文泄露
Published Aug 18
Evaluated models Gemma-4-E4B-it , Gemma-4-31B-it , Qwen2.5-14B-Instruct and others11 提出 SkillClone，仅凭良性任务交互重构闭源 Agent 技能
Published Aug 5
Evaluated models DeepSeek-Flash , DeepSeek-Pro , GLM-5.1 and others5 Abstract 论文揭示了仅靠良性任务交互即可重构闭源智能体技能隐藏功能的风险，表明功能保密需要限制累积信息泄露。论文从功能保密视角系统研究了大语言模型智能体技能的行为重构威胁。
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 SigLeak，从黑盒 Agent 执行轨迹推断专有技能
Published Jul 28
Evaluated models GPT-5.4-mini , GPT-5.5 , GLM-5.2 and others4 Abstract 提出了通过良性查询和对照轨迹推断黑盒专有技能的 SigLeak 框架，揭示了智能体执行可见性带来的行为侧信道泄露风险。
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。
提出嵌入推断攻击 EIA，在黑盒 IR 中识别所用嵌入模型
Published Jul 1
Evaluated models e5-small , e5-small-v2 , e5-base and others19 Abstract EIA 用无序文档集的差异从候选中识别嵌入模型，并给出 reranker 与阈值下的计数。EIA 是在黑盒 IR 上识别所用嵌入模型的攻击。 。
Propose soft prompt defense AREA, re-anchoring attention to mitigate system prompt leakage
Published Jun 17
Evaluated models Llama-2-7B-chat-hf , Llama-3.1-8B-Instruct , Mistral-7B-Instruct-v0.3 and others8 Abstract The authors measure commercial prompt leakage, propose attention drift, and use AREA to strike a trade-off between usability and leakage resistance.The authors study system prompt leakage in real LLM applications and propose AREA, a mitigation method that does not modify weights.
提出 MEntA，用非模板查询推断文档是否属于 RAG 检索库
Published May 23
Evaluated models Phi4-14B , Llama3.1-8B , CommandR-7B and others5 提出 SWhisper，用近超声波向语音驱动 LLM 隐蔽注入越狱提示
Published Mar 14
Evaluated models Meta-Llama-3.1-8B-Instruct , Gemma-3-4B , Qwen2.5-7B-Instruct and others9 提出 NeuroFilter，用激活探针过滤智能体多轮隐私攻击
Published Jan 21
Evaluated models GPT-OSS 20B , Qwen 2.5 32B Instruct , Llama 3.3 70B and others8 提出 iMIST，用工具伪装和 PPO 迭代越狱大语言模型
Published Jan 9
Evaluated models DeepSeek-V3 , Qwen3-32B , GPT-OSS-120B Abstract iMIST 用工具伪装和在线 PPO 做黑盒越狱，Table I 中 JADES 高于同格基线。iMIST 是黑盒越狱方法：把恶意查询伪装成合法工具调用，再用在线强化学习把响应的有害性分数往上推。
Attacks arXiv 2512.20168 · Dec 23, 2025 提出双重隐写越狱 Odysseus，用图像隐写绕过多模态系统安全过滤器
Published Dec 23, 2025
Evaluated models GPT-4o-2024-08-06 , Gemini-2.0-pro , Gemini-2.0-flash and others4