Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 17 papers Another 233 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.
提出主题锚点投毒，放大众包微调后的专有指令提取
Published Sep 28
Evaluated models Qwen2.5-7B-Instruct , Qwen2.5-14B-Instruct , Llama-3.1-8B-Instruct and others4 Abstract 论文提出基于主题锚点的数据中毒攻击，证明低比例的中毒样本可在黑盒条件下隐蔽放大微调指令泄漏。
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。
提出多轮越狱框架 BLUEPRINT，用世界观模拟与心理因素诱导有害输出
Published Sep 2
Evaluated models Qwen3-Next-80B , DeepSeek-V3.2 , GLM-4.7 and others6 Abstract BLUEPRINT 解耦多轮越狱策略与情境模拟，揭示任务具象化是摆脱拒绝的核心机制，为多轮安全防御提供新视角。BLUEPRINT 是一个将因子化社会影响力策略空间与跨轮世界观模拟解耦的多轮大语言模型安全评测框架。
提出 SMT 框架，用模拟审核轨迹越狱函数调用 LLM
Published Jul 1
Evaluated models GPT-4o , GPT-5.4 , Qwen3-Max and others7 提出 Persona Attack，分步写入对话记忆指令以越狱
Published May 29
Evaluated models GPT-4o , Llama-3.2-3B-Instruct , ChatGPT-4o and others5 提出多轮意图欺骗越狱 iDecep，用良性意图伪装引出有害内容
Published Apr 27
Evaluated models GPT-4o , Gemini-2.5-Flash , Claude-Sonnet-4.5 and others4 Abstract iDecep 用意图欺骗做多轮越狱，在 GPT-5 与 Claude-Sonnet-4.5 上同时测到直接成功和 para 成功。iDecep 是一种黑盒多轮越狱：用表面良性的意图维持与恶意目标贴近的对话，并单独计入 safe completion 替代内容中的有害信息。 。
Proposes the STAR multi-turn role-playing framework, which bypasses safety alignment through dialogue state evolution
Published Mar 15
Evaluated models GPT-4o , Claude 3.5 Sonnet , Gemini 2.0-Flash and others5 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 JustAsk，用在线技能选择提取模型系统提示词
Published Jan 29
Evaluated models anthropic/claude-opus-4.5 , openai/gpt-5.2-codex , GPT-5.2 and others19 提出 ICON 多轮越狱，用意图-上下文先验生成攻击序列
Published Jan 28
Evaluated models Llama-4-Maverick-Instruct , Llama-3.1-405B-Instruct , Qwen-Max-2025-01-25 and others11 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。
提出知识驱动多轮越狱 Mastermind，用策略知识闭环攻击对齐模型
Published Jan 9
Evaluated models Llama 3.1 8B Instruct , Llama 3.3 70B Instruct , Qwen 2.5 7B Instruct and others16 提出多轮多模态越狱，并用 FragGuard 拦截有害响应
Published Jan 9
Evaluated models LLaVa-1.6 (7B) , LLaVa-1.6 (13B) , Qwen-2-7B and others5 提出无外部攻击模型的多轮自我越狱方法 SLIP
Published Jan 6
Evaluated models GPT-5.1 , GPT-4o , Claude-Sonnet-4.5 and others11 Abstract 论文提出了无外部攻击模型的自我越狱方法 SLIP，指出对齐模型内在能力抑制的漏洞并设计了语义漂移防御。
Attacks arXiv 2512.18755 · Dec 21, 2025 提出多轮越狱框架 MEEA，用曝光递进提示与模拟退火绕过对齐
Published Dec 21, 2025
Evaluated models GPT-4 , Claude-3.5-Sonnet , DeepSeek-R1 and others5 Attacks arXiv 2511.13892 · Nov 18, 2025 提出图像排版操纵与多轮提示越狱，绕过交通场景视觉语言模型安全对齐
Published Nov 18, 2025
Evaluated models LLaVa-1.6 , Qwen-2 , GPT-4o-mini Attacks arXiv 2509.25624 · Sep 30, 2025 提出 STAC 框架，把恶意目标拆成多轮看似无害的工具调用链
Published Sep 30, 2025
Evaluated models GPT-4.1-2025-04-14 , GPT-4.1-mini-2025-04-14 , Qwen3-32B and others8 Abstract 作者提出并验证了智能体在序列化工具攻击下的普遍脆弱性，指出针对累积动作序列防御的必要性。
You have reached the end