Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 24 papers

Filter3

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 265papers have no tags yet and are not included in filtered results.

Impact：Output manipulation System component：Permissions and sandboxing System component：Fine-tuning and open weights Clear all

Another 265papers do not have category tags yet and are not included in filtered results.

AttacksarXiv 2610.12644Oct 9
Cancerous backdoor attack: progressive activation via invisible tokens and a stage controller.
Proposes the Cancerous backdoor, which progressively corrupts reasoning using invisible stage tokens.
arXiv
2610.12644
Published
Oct 9
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
Llama-3.1-8B-Instruct, Qwen2.5-7B-Instruct, Mistral-7B-Instruct-v0.3
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentReasoning chains
AbstractHidden tokens plus a stage controller make the backdoor progressively corrupt reasoning upon triggering, and the authors believe that looking only at the answer will miss it.
Cancerous Backdoor Attack changes LLM backdoors from one-time activation to multi-stage reasoning corruption that accumulates progressively with triggers.
Paper analysisRead analysis
DefensesarXiv 2610.09941Oct 7
OrthoPurify 用一步正交投影净化被后门污染的视觉语言模型
提出 OrthoPurify，用正交投影清除视觉语言模型适配器中的劫持方向
arXiv
2610.09941
Published
Oct 7
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
LLaVA-1.5-7B, Qwen3-VL-8B, LLaVA-1.5-13B and others5
- DefensesModel hardening
- AttacksPoisoning and backdoors
- ComponentFine-tuning and open weights
Abstract一次投影去掉劫持方向。
OrthoPurify 是部署前的 LVLM 后门净化方法，用一次正交投影去掉被劫持的权重更新方向。
Paper analysisRead analysis
AttacksarXiv 2610.05793Oct 5
研究者训练出按多智能体拓扑触发的代码后门模型
提出拓扑条件后门，使代码模型在推断多智能体部署时植入漏洞
arXiv
2610.05793
Published
Oct 5
Layer
Training and data layer
Setting
coding agent
Evaluated models
Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentFine-tuning and open weights
Abstract论文证实了拓扑条件后门的可训练性与泛化性，展示了常规评测盲区，并提出差异化拓扑审计作为检测防御方向。
Paper analysisRead analysis
DefensesarXiv 2610.04699Oct 3
CoVer：用干预检验为 Agent 构造可判定规则边界
提出 COVER，用干预门判定 Agent 规则谓词能否自动执行
arXiv
2610.04699
Published
Oct 3
Layer
Application layer
Setting
AI Agent
Evaluated models
Claude Sonnet, Claude Opus, GPT-4o and others5
- DefensesPermissions and approval
- ComponentPermissions and sandboxing
Abstract自可判定不能向模型索取。
这篇工作测的是智能体验证器在检查之前的分流：外部规则里的哪条谓词能用固定过程解决，哪条必须交给评审。
Paper analysisRead analysis
DefensesarXiv 2610.00685Oct 1
研究者提出用零空间投影净化 LoRA 微调 LLM 的后门
用零空间投影净化 LoRA 微调模型后门并保留基座与下游能力
arXiv
2610.00685
Published
Oct 1
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
LLaMA2-7B-Chat, LLaMA2-13B-Chat, Qwen2-7B-Instruct and others6
- DefensesModel hardening
- AttacksPoisoning and backdoors
- ComponentFine-tuning and open weights
Paper analysisRead analysis
DefensesarXiv 2609.39548Sep 30
NDDL：用正常扩散动力学为文生图模型做后门防御
提出 NDDL，用良性扩散轨迹检测文生图后门提示并定位触发 token
arXiv
2609.39548
Published
Sep 30
Layer
Model layer
Setting
Models and APIs
Evaluated models
Stable Diffusion v1.5, Stable Diffusion XL, Stable Diffusion v3.5 and others4
- DefensesMonitoring and auditing
- AttacksPoisoning and backdoors
- ComponentImage generation
AbstractNDDL 从良性扩散轨迹学习正常转移，用动态不一致检测后门并定位触发 token。
NDDL 是面向文生图扩散模型的后门防御：防御者只有白盒模型访问和良性提示词，没有后门先验，需要区分后门提示词并找出触发 token。
Paper analysisRead analysis
AttacksarXiv 2609.39047Sep 30
BadAction：通过动作引导触发器对交互式视频生成模型实施后门攻击
提出 BadAction，通过动作触发器给交互式视频生成植入后门
arXiv
2609.39047
Published
Sep 30
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
Astra (Wan2.1-T2V-1.3B)
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentFine-tuning and open weights
Paper analysisRead analysis
DefensesarXiv 2609.38909Sep 30
Purdue 提出 Pact：将欺骗作为行为遗忘
提出 PACT，遗忘推理模型在压力下的欺骗行为并保留上下文理解
arXiv
2609.38909
Published
Sep 30
Layer
Training and data layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
DeepSeek-R1-Distill-Qwen-32B, QwQ-32B
- DefensesModel hardening
- AttacksModel tampering
- RiskDeception and scheming
- ComponentFine-tuning and open weights
Abstract论文将大模型欺骗形式化为上下文条件行为，提出 PACT 框架在消除欺骗的同时保留了上下文遵循能力。
Paper analysisRead analysis
DefensesarXiv 2610.00348Sep 30
NEEDLE：通过权重正交化移除 LLM 后门
提出 NEEDLE，用权重正交化在已知触发器时去除大模型后门
arXiv
2610.00348
Published
Sep 30
Layer
Model layer
Setting
Models and APIs
Evaluated models
Gemma-3-4B-IT, Qwen3-4B-Instruct-2507, Gemma-3-12B-IT
- DefensesModel hardening
- AttacksPoisoning and backdoors
- ComponentFine-tuning and open weights
AbstractNEEDLE 用闭式权重编辑去掉已知触发器的后门，并保住拒绝子空间。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.37501Sep 28
RegLLM：面向受监管智能体 AI 的有限自主性诊断测试框架
提出 REGLLM 框架，评测受监管智能体何时作答或升级
arXiv
2609.37501
Published
Sep 28
Layer
Application layer
Setting
AI Agent
Evaluated models
Qwen2.5-3B
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
AbstractREGLLM 把升级对错做成可核验信号，并在 smoke 规模上看到适配器方向被运行方差淹没。
REGLLM 是诊断有界自主的框架，示例是 UK FCA 合规智能体。作者把它做成 smoke 规模演示，而不是完整基准验证。
Paper analysisRead analysis
AttacksarXiv 2609.37468Sep 26
论文提出针对 Agentic RAG 的检索器后门攻击与 DOU 隐藏方法
提出检索器后门攻击操纵智能体搜索，并用 DOU 逃避检测
arXiv
2609.37468
Published
Sep 26
Layer
Application layer
Setting
AI Agent
Evaluated models
E5, Contriever-MSMARCO
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentRAG
Abstract论文展示了通过恶意检索器操纵智能体多轮搜索并利用诱饵重塑表征隐蔽后门的可行性与系统风险。
Paper analysisRead analysis
DefensesarXiv 2609.31282Sep 25
基于区块链的 Agentic AI 框架：面向安全软件供应链的资源优化与能耗感知
提出 AttestChain，将供应链安全智能体报告签名上链并据此控制发布
arXiv
2609.31282
Published
Sep 25
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
GPT-class LLM
- DefensesWatermarking and provenance
- ComponentPermissions and sandboxing
AbstractAttestChain 将智能体报告与链上发布门控结合，现有证据来自受控合成构建的总体结果。
AttestChain 是一个把 LLM 安全检查、智能体身份验证和许可链发布策略连接起来的软件供应链防御框架。。
Paper analysisRead analysis
AttacksarXiv 2609.27422Sep 23
RAMP：逆转对抗扰动以增强针对恶意软件检测器的干净标签后门攻击
提出 RAMP，逆转对抗扰动加强恶意软件检测器的干净标签后门
arXiv
2609.27422
Published
Sep 23
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
MalConv, MalConvGCG
- AttacksPoisoning and backdoors
- TechniqueAutomated search
- ComponentFine-tuning and open weights
AbstractRAMP 用反转扰动做 clean-label 投毒。
RAMP 是 score-based black-box 下的 clean-label 后门增强，对象是会微调的 Windows PE 原始字节恶意软件检测器。
Paper analysisRead analysis
DefensesarXiv 2609.11757Sep 11
Researchers disclose a Whisper prompt injection attack on the AP2 payment protocol and propose the A-VIP defense
提出 AP2 支付协议的 Whisper 攻击与 A-VIP 绑定防御
arXiv
2609.11757
Published
Sep 11
Layer
Application layer
Setting
AI Agent
Evaluated models
gemini-2.5-flash-lite, gemini-3.1-flash-lite, gemini-3-flash-preview and others15
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Abstract论文揭示了 AP2 协议决策层受操纵的风险，提出结构化与权能绑定防御 A-VIP，并发布评测基准。
Paper analysisRead analysis
AttacksarXiv 2609.07147Sep 7
FDBA：联邦学习中的细粒度分布式后门攻击
提出联邦学习细粒度分布式后门框架 FDBA，用动态触发与对比嵌入降低投毒比例
arXiv
2609.07147
Published
Sep 7
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
ResNet-18
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentFine-tuning and open weights
Paper analysisRead analysis
DefensesarXiv 2609.04159Sep 4
Sentinel-RL：将拓扑推理从 LLM 智能体中卸载到安全运营中心
提出 Sentinel-RL，将拓扑处置从 LLM 卸载到受约束图策略
arXiv
2609.04159
Published
Sep 4
Layer
Application layer
Setting
LLM applications
Evaluated models
HetGAT, PPO
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
AbstractSentinel-RL 用图编码器与 PPO 承担拓扑决策，LLM 负责叙述与门控，并在 LANL 上报告检测与时延。
Sentinel-RL 是面向 SOC 的神经符号架构：拓扑决策离开 LLM，改由认证图上的 HetGAT 与 PPO 完成。
Paper analysisRead analysis
AttacksarXiv 2609.04260Sep 2
GhostWord：针对自动语音识别的细粒度后门攻击
提出 GhostWord 词级触发后门，对自动语音识别做数据投毒
arXiv
2609.04260
Published
Sep 2
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
Whisper-Small, Whisper-Medium, MMS and others4
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentAudio
Paper analysisRead analysis
DefensesarXiv 2609.01091Sep 1
上海交大等提出 trait-direction drift 机制与探针空间走廊正则，抑制蒸馏中的潜隐特质迁移
提出特征方向漂移机制与探针空间走廊正则，抑制蒸馏潜隐特质迁移
arXiv
2609.01091
Published
Sep 1
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
Qwen2.5-7B-Instruct, Llama-3.1-8B-Instruct, Gemma-3-12B-IT
- DefensesModel hardening
- AttacksPoisoning and backdoors
- RiskMisalignment and value divergence
- ComponentFine-tuning and open weights
Abstract论文将潜意识学习归结为特征方向漂移的持续累积，并提出走廊正则化在微调中控制隐蔽特征转移。
Paper analysisRead analysis
OtherarXiv 2608.30387Aug 31
研究者提出多智能体系统输出与委派链路双层存证方案
提出多智能体输出与跨部署方委派的两层存证
arXiv
2608.30387
Published
Aug 31
Layer
Application layer
Setting
Multi-agent systems
- ComponentPermissions and sandboxing
Abstract两层证明记录放出字节与边授权。
部署方侧的两层证明，用来在多智能体跨部署方委托之后回答两个事后问题：谁放出了这些字节，这条跨部署方边有没有父方授权。
Paper analysisRead analysis
Analysis and theoryarXiv 2609.00060Aug 30
论文用 Tamarin 形式化分析 x402、MPP、ACP、AP2 四个智能体支付协议
用 Tamarin 形式化分析智能体支付协议的跨阶段安全关系
arXiv
2609.00060
Published
Aug 30
Layer
Application layer
Setting
AI Agent
Evaluated models
x402, MPP, ACP and others4
- ComponentPermissions and sandboxing
Paper analysisRead analysis

AI Safety Daily

Read the latest issueOne issue per day, updated as events develop

All papers

Attack type

Attack technique

Defense type

Impact

System type

Attack surface / entry point

Layer

System component

Cancerous backdoor attack: progressive activation via invisible tokens and a stage controller.

OrthoPurify 用一步正交投影净化被后门污染的视觉语言模型

研究者训练出按多智能体拓扑触发的代码后门模型

CoVer：用干预检验为 Agent 构造可判定规则边界

研究者提出用零空间投影净化 LoRA 微调 LLM 的后门

NDDL：用正常扩散动力学为文生图模型做后门防御

BadAction：通过动作引导触发器对交互式视频生成模型实施后门攻击

Purdue 提出 Pact：将欺骗作为行为遗忘

NEEDLE：通过权重正交化移除 LLM 后门

RegLLM：面向受监管智能体 AI 的有限自主性诊断测试框架

论文提出针对 Agentic RAG 的检索器后门攻击与 DOU 隐藏方法

基于区块链的 Agentic AI 框架：面向安全软件供应链的资源优化与能耗感知

RAMP：逆转对抗扰动以增强针对恶意软件检测器的干净标签后门攻击

Researchers disclose a Whisper prompt injection attack on the AP2 payment protocol and propose the A-VIP defense

FDBA：联邦学习中的细粒度分布式后门攻击

Sentinel-RL：将拓扑推理从 LLM 智能体中卸载到安全运营中心

GhostWord：针对自动语音识别的细粒度后门攻击

上海交大等提出 trait-direction drift 机制与探针空间走廊正则，抑制蒸馏中的潜隐特质迁移

研究者提出多智能体系统输出与委派链路双层存证方案

论文用 Tamarin 形式化分析 x402、MPP、ACP、AP2 四个智能体支付协议