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

Oct 12 · Monday

All papers

Found 8 papers

Filter3

Attack type

Jailbreaks15 Poisoning and backdoors3 Extraction and theft5 Detection evasion7 Malicious use7

Impact

Data and privacy leaks5

System type

Models and APIs8

Attack surface / entry point

User input3 Weights and gradients5 Distribution channels3

Layer

Model layer5

System component

MCP and skills44 Memory43 RAG29 Guardrails and judges12 Monitors7 Reasoning chains10 Permissions and sandboxing13 Fine-tuning and open weights33 Vision14 Audio5 Image generation8 Embeddings9 Multimodal28

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 232papers have no tags yet and are not included in filtered results.

Attack type：Poisoning and backdoors Attack type：Extraction and theft System component：Image generation Clear all

Another 232papers do not have category tags yet and are not included in filtered results.

AttacksarXiv 2610.05601Oct 5
Tracer 框架可从遗忘后的扩散模型中识别被擦除概念
提出 TRACER，从权重差识别扩散模型被擦除概念
arXiv
2610.05601
Published
Oct 5
Layer
Model layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
Stable Diffusion v1.5, Stable Diffusion 3, FLUX.1 and others5
- AttacksExtraction and theft
- ComponentImage generation
AbstractTRACER 用权重足迹识别未披露擦除概念并估计个数，作者称数秒内完成且准确率高于对照。
TRACER 针对的是遗忘扩散模型上的目标识别：提供者从公开基座 W 用已发表方法擦掉未知集合 S⋆，只发布 W′。攻击者还有假定包含这些概念的词表，但不知擦了谁、擦了几个、用了什么算法。
Paper analysisRead analysis
AttacksarXiv 2610.05453Oct 5
研究者提出 Privacy-Leaking Watermarks，可在本地多模态模型中泄露对话隐私
提出 PLW，用微调把先前对话敏感触发编码进后续生成图像
arXiv
2610.05453
Published
Oct 5
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
OmniGen2, BAGEL-7B
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentFine-tuning and open weights
AbstractPLW 把先前对话线索写成后续图像中的隐水印，检出随触发、模型和间隔变化。
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 2609.23586Sep 20
VidMark：面向视频生成模型知识产权保护的高效水印方案
提出 VidMark 水印方案，核验视频生成模型的输出来源与所有权
arXiv
2609.23586
Published
Sep 20
Layer
Model layer
Setting
Models and APIs
Evaluated models
Stable Video Diffusion (SVD), Open-Sora, Wan
- DefensesWatermarking and provenance
- AttacksDetection evasion
- ComponentImage generation
AbstractVidMark 加解码器引导微调，把模型签名写入 VGM 输出并用于两项核验。
Huang 等人提出一套面向视频生成模型的生成内嵌水印方案，用来同时判断一条视频是否来自受保护模型，以及一个嫌疑模型是否为该模型的拷贝。
Paper analysisRead analysis
DefensesarXiv 2609.15799Sep 15
RAPID：面向文生图服务的实时防未授权蒸馏防御
提出 RAPID，离线把抗蒸馏扰动写入文生图解码器
arXiv
2609.15799
Published
Sep 15
Layer
Model layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
PixArt, SSD, SDXL and others4
- DefensesModel hardening
- AttacksExtraction and theft
- ComponentImage generation
Paper analysisRead analysis
AttacksarXiv 2509.09488Sep 11, 2025
研究者提出 SeedSnitch 与 PromptPirate，利用种子漏洞窃取扩散模型提示词
提出 SeedSnitch 与 PromptPirate，恢复种子后窃取扩散提示词
arXiv
2509.09488
Published
Sep 11, 2025
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
Stable Diffusion 3.5 Large, Stable Diffusion 3.5 Medium, Stable Diffusion 3.5 Turbo and others4
- AttacksExtraction and theft
- TechniqueAutomated search
- ComponentImage generation
Paper analysisRead analysis
AttacksarXiv 2508.06837Aug 9, 2025
Prometheus：针对文生图扩散模型的免训练提示词窃取攻击
提出 Prometheus，从展示图反推文生图模型的在售提示词
arXiv
2508.06837
Published
Aug 9, 2025
Layer
Model layer
Setting
Models and APIs
Attacker
No-box (offline transfer)
Evaluated models
Midjourney, Leonardo.ai, DALL·E and others4
- AttacksExtraction and theft
- TechniqueAutomated search
- ComponentImage generation
Abstract训练无关的 proxy 搜索攻击。
Prometheus 是针对文生图市场 showcase 的训练无关提示词窃取：攻击者看不到在售文本，只用本地开源 proxy 恢复 subject 和 modifier，并希望换 subject 后风格仍在。
Paper analysisRead analysis

You have reached the end

AI Safety Daily

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