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

Oct 12 · Monday

All papers

Found 105 papers

Filter2

System type

Models and APIs80 LLM applications14 AI Agent12

Attacker access

Black-box27 Gray-box6 White-box26

Attack surface / entry point

User input30 Documents and messages13 Tool responses4 Tool and skill descriptions4 Images, audio and video24 Training data18 Weights and gradients33 Distribution channels5 Inference services and hardware3

Layer

Prompt layer22 Model layer39 Application layer23 Training and data layer20

System component

MCP and skills4 RAG3 Guardrails and judges22 Monitors6 Reasoning chains7 Fine-tuning and open weights22 Vision8 Video3 Image generation8 Multimodal19

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 321papers have no tags yet and are not included in filtered results.

Attack type：Model tampering Attack type：Detection evasion Clear all

Another 321papers do not have category tags yet and are not included in filtered results.

AttacksarXiv 2610.12292Oct 9
The paper proposes an option channel attack: changing one option label can make an agent guardrail allow a prohibited operation
Proposing an option channel attack that renames option labels so that agent guardrails allow prohibited operations
arXiv
2610.12292
Published
Oct 9
Layer
Application layer
Setting
LLM applications
Attacker
Black-box
Evaluated models
LAYA-TD, LAYA-EN, LAYA-ML and others5
- AttacksDetection evasion
- ComponentGuardrails and judges
AbstractThe authors' evaluation indicates that typed models should not serve as final safety arbiters; they are susceptible to off-policy text and option name manipulation, and confidence filtering cannot resist attacks.
Paper analysisRead analysis
DefensesarXiv 2610.11848Oct 8
SemField: training-free semantic watermark
Proposes the semantic watermark SemField, which still identifies the attribution of generated text after paraphrasing and structural changes.
arXiv
2610.11848
Published
Oct 8
Layer
Model layer
Setting
Models and APIs
Evaluated models
Llama-3.2-1B, Qwen3-1.7B, OPT-1.3B
- DefensesWatermarking and provenance
- AttacksDetection evasion
AbstractTraining-independent document-level semantic watermark.
SemField is a zero-bit semantic watermark embedded by the provider at generation time and detected later with a key, used to identify the provider's own model text after paraphrasing and sentence-level structural changes.
Paper analysisRead analysis
AttacksarXiv 2610.10742Oct 8
BRANCH 方法绕过 6 套多扫描器护栏系统
提出 BRANCH 方法，用分支搜索绕过多扫描器护栏
arXiv
2610.10742
Published
Oct 8
Layer
Application layer
Setting
LLM applications
Attacker
Black-box
Evaluated models
Meta Llama Guard 3 8B, Meta Llama Guard 7B, Meta Llama Guard 3 1B and others18
- AttacksDetection evasion
- TechniqueAutomated search
- ComponentGuardrails and judges
Paper analysisRead analysis
DefensesarXiv 2610.10345Oct 8
SLDR：通过选择性层恢复与动态路由防御恶意微调
提出 SLDR，用选择性层恢复与动态路由防御恶意微调
arXiv
2610.10345
Published
Oct 8
Layer
Training and data layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
Llama3.1-8B-Instruct, Llama3-8B-Instruct, Qwen2.5-7B-Instruct and others4
- DefensesModel hardening
- AttacksModel tampering
- ComponentFine-tuning and open weights
Abstract作者称 SLDR 能压低有害输出并保持下游效用。
SLDR 是针对恶意微调的微调后防御：在已对齐模型上做有符号层探测，只在敏感分数最大与最小的两层训练 LoRA，再用 lmax 的最后 token 表示决定是否启用适配器。
Paper analysisRead analysis
DefensesarXiv 2610.09793Oct 7
用时序逻辑监控工具 Agent
提出 MFOTL 离线时序验证以监控工具 Agent 的危险动作链
arXiv
2610.09793
Published
Oct 7
Layer
Application layer
Setting
AI Agent
Attacker
Black-box
- DefensesMonitoring and auditing
- AttacksPrompt injection
- ComponentMonitors
Abstract形式化验证可有效识别多步攻击，但其有效性高度依赖轨迹的可信历史记录。
这篇论文对工具调用大语言模型智能体的安全监控进行了形式化时序验证研究。
Paper analysisRead analysis
AttacksarXiv 2610.10612Oct 7
PyCache Trap 利用 Python 缓存替换绕过七款 Agent Skill 扫描器
提出 PyCache Trap，用字节码缓存替换绕过 Agent 技能扫描器
arXiv
2610.10612
Published
Oct 7
Layer
Application layer
Setting
coding agent
Evaluated models
CISCO, HSS, AUDITOR and others8
- DefensesDetection and filtering
- AttacksDetection evasion
- TechniqueDefense-targeted
- ComponentMCP and skills
Abstract论文剖析智能体技能审查与执行脱节盲区，提出 PyCache Trap 攻击并构建基于可信复现的 EAV 防御。
这篇论文研究了智能体技能（Agent Skills）安全扫描中的“审查–执行脱节”问题，并提出了对应的攻击验证与防御方案。
Paper analysisRead analysis
Analysis and theoryarXiv 2610.09004Oct 7
研究：去除有害信息不足以证明开源权重模型的抗篡改能力
论证发布时互信息不能认证开源权重抗微调恢复
arXiv
2610.09004
Published
Oct 7
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
EleutherAI/deep-ignorance-e2e-strong-filter, EleutherAI/deep-ignorance-unfiltered, Google BERT-Tiny and others6
- DefensesModel hardening
- AttacksModel tampering
- ComponentFine-tuning and open weights
Abstract作者称零信息仍可一步恢复，表观抗微调可来自参数化。
作者研究开源权重模型的篡改抵抗：发布时的互信息是否足以保证微调恢复很慢。
Paper analysisRead analysis
InterpretabilityarXiv 2610.09000Oct 7
研究定位 LLaMA-2-7B-Chat 安全敏感参数
定位安全敏感参数并稀疏改动权重以削弱安全对齐
arXiv
2610.09000
Published
Oct 7
Layer
Model layer
Setting
Models and APIs
Evaluated models
LLaMA-2-7B-Chat
- AttacksModel tampering
- RiskInappropriate refusal
Paper analysisRead analysis
DefensesarXiv 2610.08668Oct 7
SBW：面向 LLM 智能体的语义行为水印方法
提出 SBW，为 LLM Agent 轨迹做抗改写与伪造的语义行为水印
arXiv
2610.08668
Published
Oct 7
Layer
Application layer
Setting
AI Agent
Evaluated models
Qwen2.5-3B-Instruct, LLaMA-3.2-3B, Qwen2.5-7B and others5
- DefensesWatermarking and provenance
- AttacksDetection evasion
AbstractSBW 在语义簇上做溯源水印，改写检测高于精确符号，链式重放仍开放。
SBW 是免训练的智能体行为水印，把所有者标识放在离散动作选择上，而不是输出 token。
Paper analysisRead analysis
AttacksarXiv 2610.08331Oct 6
STCA：研究者提出针对自动驾驶视频 VLM 的时空一致性对抗攻击
提出 STCA，对驾驶视频 VLM 构造可迁移的时空对抗扰动
arXiv
2610.08331
Published
Oct 6
Layer
Prompt layer
Setting
Models and APIs
Attacker
No-box (offline transfer)
Evaluated models
Video-LLaVA-7B, Qwen2.5-VL-7B, Dolphins
- AttacksDetection evasion
- TechniqueGradient and representation optimization
- ComponentVideo
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.07939Oct 6
CCDF 数据集：Grok Imagine、Veo 3.1、Sora 2 伪造监控录像，十个深伪检测器基本认不出
提出监控深伪基准 CCDF，在统一编码后评测检测器对伪造犯罪事故录像的识别
arXiv
2610.07939
Published
Oct 6
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
C2P-CLIP, ForgeLens, D3-Im and others10
- DefensesDetection and filtering
- AttacksDetection evasion
- ComponentGuardrails and judges
Abstract提出消除格式捷径的监控深度伪造基准 CCDF，揭示现有检测器难以识别商用生成视频。
Paper analysisRead analysis
InterpretabilityarXiv 2610.07518Oct 6
TRACE：仅凭 checkpoint 更新审计大语言模型有害目标
提出 TRACE，仅凭检查点更新审计有害合规微调目标
arXiv
2610.07518
Published
Oct 6
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
Llama-3.1-8B-Instruct, Mistral-7B-Instruct-v0.3, Qwen3-8B and others6
- AttacksModel tampering
- ComponentFine-tuning and open weights
Abstract作者称有害合规 SFT 在检查点更新中留下可排序坐标，并据此做只读权重审计。
作者在检查点更新里寻找有害合规 SFT 的目标组成，并把读数做成不查询模型、不看未知训练数据的审计分 TRACE。
Paper analysisRead analysis
AttacksarXiv 2610.07323Oct 6
ATLAS：用自适应信任域与主动学习搜索潜在对抗样本集
提出 ATLAS 用水平集估计恢复黑盒分类器的对抗输入集合
arXiv
2610.07323
Published
Oct 6
Layer
Prompt layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
LeNet5, ResNet50 (CIFAR-10 standard), ResNet50 (ImageNet standard) and others5
- AttacksDetection evasion
- TechniqueAutomated search
AbstractATLAS 用主动水平集估计构造黑盒对抗集，作者称其集合比单点攻击更具代表性。
ATLAS 是一个查询式 black-box 框架，用来构造诱发失败的对抗输入集合，供鲁棒性审计使用。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.06339Oct 5
BabelFake：面向多语言音视频 DeepFake 检测的基准数据集
构建多语言音视频 DeepFake 基准 BabelFake 并评测检测器
arXiv
2610.06339
Published
Oct 5
Layer
Model layer
Setting
Models and APIs
Evaluated models
SpeechForensics, AVH-Align, (B)RAVEn
- DefensesDetection and filtering
- AttacksDetection evasion
- ComponentAudio
AbstractBabelFake 为五语同意采集基准。
BabelFake 是经同意与 IRB 批准采集的多语音视频 DeepFake 检测基准，用来把视觉操纵和语音合成拆开比较。
Paper analysisRead analysis
AttacksarXiv 2610.05894Oct 5
研究者提出闭环偏见注入攻击
提出闭环激活引导，在扩散语言模型去噪过程中注入定向偏见
arXiv
2610.05894
Published
Oct 5
Layer
Model layer
Setting
Models and APIs
Attacker
Gray-box
Evaluated models
LLaDA-8B-Instruct, Dream-v0-Instruct-7B, LLaDA-MoE-7B-A1B-Instruct
- AttacksModel tampering
- TechniqueGradient and representation optimization
Abstract通过去噪中间概率反馈动态调节残差流转向，证明扩散语言模型在推理服务栈中存在闭环偏见注入脆弱性。
Paper analysisRead analysis
DefensesarXiv 2610.05870Oct 5
MBZUAI 提出可校准的真实图像认证方法
提出校准重合成方法，以生成器反演保真度认证真实图像并防御对抗扰动
arXiv
2610.05870
Published
Oct 5
Layer
Application layer
Setting
LLM applications
Attacker
White-box
- DefensesDetection and filtering
- AttacksDetection evasion
- ComponentImage generation
Abstract论文提出基于反演保真度与阈值校准的健全认证框架，在对抗扰动下维持错误率界限，并指出真实内容可认证性随生成器演进而萎缩。
Paper analysisRead analysis
DefensesarXiv 2610.04763Oct 4
UT Austin 将 SynthID 水印嵌入基因组语言模型，凭 DNA 序列和密钥追溯 AI 生成片段
把 SynthID 嵌入基因组语言模型采样，并用位置无关检测追溯生成序列
arXiv
2610.04763
Published
Oct 4
Layer
Model layer
Setting
Models and APIs
Evaluated models
Carbon-500M, GENERator-v2-eukaryote-1.2b-base
- DefensesWatermarking and provenance
- AttacksDetection evasion
Abstract两个六碱基基因组模型嵌入 SynthID 后，位置无关验证能找回水印，作者称这只是统计检测。
作者在两个六碱基基因组模型的采样里嵌入 SynthID，并让验证者在不知道起点、链和 token phase 时做一次校正后的检出。
Paper analysisRead analysis
DefensesarXiv 2610.04235Oct 3
Thrive：面向自回归 TTS 的抗重建攻击多比特语音水印框架
提出 THRIVE，为自回归 TTS 嵌入抗重建的多比特语音水印
arXiv
2610.04235
Published
Oct 3
Layer
Model layer
Setting
Models and APIs
Evaluated models
Qwen3-TTS (0.6B), Dots.tts (2B)
- DefensesWatermarking and provenance
- AttacksDetection evasion
- ComponentAudio
Paper analysisRead analysis
AttacksarXiv 2610.03166Oct 2
LiBRA：通过双向隐空间优化实现检测感知的图像水印去除
提出 LiBRA，在潜空间双向优化以规避图像水印双侧检测
arXiv
2610.03166
Published
Oct 2
Layer
Model layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
Stable Signature, YU1, HiDDeN and others4
- AttacksDetection evasion
- TechniqueDefense-targeted
AbstractLiBRA 用双向潜空间优化规避双侧水印检测，并提高相对反转基线的保真。
Paper analysisRead analysis
DefensesarXiv 2610.02853Oct 2
研究提出基于收缩约束状态空间模型的有界可达性与越狱检测方法
提出收缩约束 SSM 安全头，认证嵌入扰动并检测越狱
arXiv
2610.02853
Published
Oct 2
Layer
Prompt layer
Setting
Models and APIs
Attacker
White-box, Black-box
Evaluated models
ToySSMClassifier, S4 safety head
- DefensesDetection and filtering
- AttacksJailbreaks
- ComponentGuardrails and judges
Abstract收缩给出玩具 LTI 的有界认证。
给 SSM 安全头一个收缩条件，并在越狱检测上使用收缩正则的 S4 头。。
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

Attacker access

Attack surface / entry point

Layer

System component

The paper proposes an option channel attack: changing one option label can make an agent guardrail allow a prohibited operation

SemField: training-free semantic watermark

BRANCH 方法绕过 6 套多扫描器护栏系统

SLDR：通过选择性层恢复与动态路由防御恶意微调

用时序逻辑监控工具 Agent

PyCache Trap 利用 Python 缓存替换绕过七款 Agent Skill 扫描器

研究：去除有害信息不足以证明开源权重模型的抗篡改能力

研究定位 LLaMA-2-7B-Chat 安全敏感参数

SBW：面向 LLM 智能体的语义行为水印方法

STCA：研究者提出针对自动驾驶视频 VLM 的时空一致性对抗攻击

CCDF 数据集：Grok Imagine、Veo 3.1、Sora 2 伪造监控录像，十个深伪检测器基本认不出

TRACE：仅凭 checkpoint 更新审计大语言模型有害目标

ATLAS：用自适应信任域与主动学习搜索潜在对抗样本集

BabelFake：面向多语言音视频 DeepFake 检测的基准数据集

研究者提出闭环偏见注入攻击

MBZUAI 提出可校准的真实图像认证方法

UT Austin 将 SynthID 水印嵌入基因组语言模型，凭 DNA 序列和密钥追溯 AI 生成片段

Thrive：面向自回归 TTS 的抗重建攻击多比特语音水印框架

LiBRA：通过双向隐空间优化实现检测感知的图像水印去除

研究提出基于收缩约束状态空间模型的有界可达性与越狱检测方法