Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 254papers have no tags yet and are not included in filtered results.

Attack type：Jailbreaks Attack type：Extraction and theft Attack surface / entry point：Images, audio and video Clear all

Another 254papers do not have category tags yet and are not included in filtered results.

DefensesarXiv 2610.09703Oct 7
研究揭示 Token-Pruning 对 VLM 安全性的影响并提出 SAP 缓解机制
提出 Safety-Aware Pruning，缓解视觉 token 剪枝放大的多模态越狱
arXiv
2610.09703
Published
Oct 7
Layer
Model layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
LLaVA-1.5-7B
- DefensesInference-time intervention
- AttacksJailbreaks
- ComponentVision
Abstract剪枝去掉背景 token 后注意力塌向恶意前景。
这篇工作评估视觉 token 剪枝如何改变 VLM 在多模态越狱下的行为，并给出推理时防御 SAP。
Paper analysisRead analysis
Analysis and theoryarXiv 2610.07009Oct 4
研究受控拆解图像到文本越狱中图像属性的作用
固定有害指令，单轴归因图像到文本越狱的图像属性
arXiv
2610.07009
Published
Oct 4
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
GPT-4o, Claude Sonnet 4.5, Qwen3-VL-8B-Instruct and others6
- AttacksJailbreaks
- ComponentVision
Abstract固定文本的图像越狱拆解，只在 Qwen 上给出一条有界相关性归因。
Paper analysisRead analysis
Analysis and theoryarXiv 2610.07005Oct 4
研究审计 Qwen2-Audio 音频越狱 AdvWave-P 的频率与解码层归因
遮挡审计 AdvWave-P 音频越狱的频带必要性与解码层关联
arXiv
2610.07005
Published
Oct 4
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
Qwen2-Audio-7B-Instruct, LLaMA-Omni
- AttacksJailbreaks
- ComponentAudio
AbstractAdvWave-P 的八频带排序与能量相关。
这是对加性音频越狱 AdvWave-P 的频率与解码器深度审计，受害模型主要是 Qwen2-Audio-7B-Instruct，不提出新攻击。
Paper analysisRead analysis
AttacksarXiv 2610.01871Oct 1
研究者提出 IMMRAG 攻击，可从多模态 RAG 数据存储中提取医学与文档图像
提出 imMRAG 攻击，从多模态 RAG 私有图像库提取内容
arXiv
2610.01871
Published
Oct 1
Layer
Application layer
Setting
LLM applications
Attacker
Black-box
Evaluated models
Lumina-mGPT, Gemini 2.5 Flash Image Preview, CLIP ViT-B/16 and others5
- AttacksExtraction and theft
- TechniqueCross-modal payloads
- ComponentRAG
Paper analysisRead analysis
AttacksarXiv 2609.38899Sep 30
SceneJail: Jailbreaking multimodal large models using video scene context
提出 SceneJail，利用视频情境上下文越狱视频多模态模型
arXiv
2609.38899
Published
Sep 30
Layer
Prompt layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
Qwen2.5-VL-7B-Instruct, Qwen2.5-VL-32B-Instruct, Qwen3-VL-8B-Instruct and others8
- AttacksJailbreaks
- TechniqueCross-modal payloads
- ComponentVideo
Paper analysisRead analysis
DefensesarXiv 2609.36562Sep 29
ThinkingGuard：通过逐步风险归因解码多模态大语言模型中的隐式危害
构建 TriggerBench 并训练 ThinkingGuard，检测图文隐式风险
arXiv
2609.36562
Published
Sep 29
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
ThinkingGuard（Qwen3-VL-8B-Instruct）
- DefensesDetection and filtering
- AttacksJailbreaks
- ComponentGuardrails and judges
AbstractTriggerBench 隔离触发条件，ThinkingGuard 用分步归因在隐式风险检测上取得作者报告的较高平均分。
这篇工作研究多模态隐式风险检测：图文各自良性，风险只在关键元素与触发元素交互时出现。
Paper analysisRead analysis
Analysis and theoryarXiv 2609.33898Sep 28
研究揭示训练效率与模型脆弱性的权衡：高效训练更易受对抗与隐私攻击
比较高效与高成本训练的对抗和隐私脆弱性并分析损失几何
arXiv
2609.33898
Published
Sep 28
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
ResNet-18, Swin Transformer, Qwen2.5-Math-7B-Instruct and others7
- AttacksExtraction and theft
Abstract效用匹配下，三种高效训练都伴随更高的隐私与对抗脆弱性及更尖的损失几何。
作者在视觉分类和数学推理语言模型上，检查用更少数据或更简对齐换取训练效率时，安全性质是否一起变化。比较在任务效用接近的前提下进行。视觉侧用 KNN-Shapley 的高重要性小子集对照低重要性大子集，并用预训练微调对照从头训练。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.33481Sep 27
IDUnlearn-Bench：面向视觉语言模型的个体级遗忘基准
提出 IDUnlearn-Bench，评测 VLM 个体级遗忘后的身份残留
arXiv
2609.33481
Published
Sep 27
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
Qwen3-VL-2B, Qwen3-VL-4B, Qwen3-VL-8B and others6
- DefensesModel hardening
- AttacksExtraction and theft
- ComponentVision
AbstractIDUnlearn-Bench 用四族探针显示：压低直接属性访问后，身份仍常能被反向查出、绑定或重构。
IDUnlearn-Bench 评测 VLM 是否忘掉一个可跨模态识别的人，而不只是答不出某条监督事实。
Paper analysisRead analysis
DefensesarXiv 2609.29287Sep 24
AEGIS：用内部信号在中间层拦截大型音频语言模型越狱
提出 AEGIS，用中层内部信号选择性激活后层适配器拦截音频越狱
arXiv
2609.29287
Published
Sep 24
Layer
Model layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
Qwen2-Audio-7B, VITA-1.5, Ultravox v0.5 and others6
- DefensesInference-time intervention
- AttacksJailbreaks
- ComponentAudio
Paper analysisRead analysis
DefensesarXiv 2609.27378Sep 23
PALS：面向全双工语音对话模型对抗鲁棒性的心理声学对齐潜空间平滑
提出 PALS，用潜空间平滑防御全双工语音对话的对抗扰动
arXiv
2609.27378
Published
Sep 23
Layer
Model layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
Moshi-style agent
- DefensesModel hardening
- AttacksJailbreaks
- ComponentAudio
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 2609.14258Sep 13
SPARK：在 KV 记忆层对齐表征以防御视觉语言模型越狱
提出 SPARK，在预填充阶段修复多模态 KV 记忆以防御视觉语言模型越狱
arXiv
2609.14258
Published
Sep 13
Layer
Model layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
LLaVA-OneVision-7B, Chameleon-7B, Qwen2-VL-7B and others4
- DefensesInference-time intervention
- AttacksJailbreaks
- ComponentVision
Paper analysisRead analysis
DefensesarXiv 2609.10613Sep 8
后验重加权框架解释并缓解多模态大模型上下文越狱
用后验重加权解释多模态上下文越狱，并提出风险门控的良性示例注入防御
arXiv
2609.10613
Published
Sep 8
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
Qwen3-VL-2B, Qwen3-VL-4B, Qwen3-VL-8B and others8
- DefensesInstruction–data separation
- AttacksJailbreaks
- ComponentVision
Abstract后验重加权解释上下文越狱的缩放，并支持风险门控的良性示例防御。
Zhang 等把多模态 ICL 越狱写成对齐模型在安全与有害隐模式之间的推理时后验漂移，并据此做风险门控防御。
Paper analysisRead analysis
AttacksarXiv 2609.07216Sep 7
DIVA：利用视觉锚点对视频生成模型实施多模态越狱
提出 DIVA，将有害意图拆成视觉锚点与运动文本以越狱视频生成模型
arXiv
2609.07216
Published
Sep 7
Layer
Prompt layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
Wan, CogVideoX, LTX-Video and others9
- AttacksJailbreaks
- TechniqueCross-modal payloads
- ComponentImage generation
AbstractDIVA 用视觉锚点抑制语义漂移做视频越狱，并给出多条件安全基准 TI2VSafetyBench。
DIVA 是针对带参考图的视频生成模型的免训练 black-box 越狱：用一致性机制本身堵住语义漂移这条安全逃逸路径。。
Paper analysisRead analysis
AttacksarXiv 2609.05525Sep 2
DIVA：针对离散扩散视觉语言模型的跨步条件传播视觉越狱
提出 DIVA，利用跨步条件传播对离散扩散视觉语言模型实施视觉越狱
arXiv
2609.05525
Published
Sep 2
Layer
Prompt layer
Setting
Models and APIs
Attacker
White-box
Evaluated models
LLaDA-V, MMaDA, LLaDA2.0-Uni
- AttacksJailbreaks
- TechniqueGradient and representation optimization
- ComponentVision
Paper analysisRead analysis
AttacksarXiv 2608.27531Aug 27
MAMJ：面向视觉语言模型的元自适应多模态越狱攻击
提出 MAMJ，在元层面交替优化多模态越狱的策略提示与攻击者权重
arXiv
2608.27531
Published
Aug 27
Layer
Prompt layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
GPT-4o, Gemini-3-Pro-Preview, Seed 2.0
- AttacksJailbreaks
- TechniqueAutomated search
- ComponentVision
Paper analysisRead analysis
AttacksarXiv 2608.15578Aug 16
ARENA：面向大型音频语言模型的自动化红队框架
提出 ARENA 音频红队框架，自动构造文本安全且音频接地的越狱输入
arXiv
2608.15578
Published
Aug 16
Layer
Prompt layer
Setting
Models and APIs
Attacker
Black-box
Evaluated models
Audio Flamingo 3, Qwen2-Audio-7B-Instruct, MiMo-Audio-7B-Instruct and others4
- AttacksJailbreaks
- TechniqueAutomated search
- ComponentAudio
Paper analysisRead analysis
AttacksarXiv 2608.04034Aug 3
HACA：原子越狱策略解耦组合的多模态自动红队方法
提出 HACA，将图文越狱拆为原子策略并自动组合以诱导有害输出
arXiv
2608.04034
Published
Aug 3
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
LLaVA-1.6-Mistral-7B, InternVL-3.5-8B, VLGuard (LLaVA-v1.5-7B-Mixed) and others5
- AttacksJailbreaks
- TechniqueCross-modal payloads
- ComponentVision
Paper analysisRead analysis
AttacksarXiv 2607.02961Jul 3
研究者提出信息过载方法越狱大型视觉语言模型
提出 INFER+ 信息过载方法，越狱大型视觉语言模型
arXiv
2607.02961
Published
Jul 3
Layer
Prompt layer
Setting
Models and APIs
Attacker
Black-box, Gray-box
Evaluated models
Qwen3-VL-8B, Qwen2-VL-7B, InternVL3.5-8B and others8
- AttacksJailbreaks
- TechniqueCross-modal payloads
- ComponentVision
Abstract用图文过载越狱 LVLM。
这篇工作用信息过载越狱 LVLM，而不是靠 OOD 变换隐藏意图。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2606.07706Jun 5
MLingualFC: Evaluating Jailbreak Vulnerabilities in Multilingual Vision-Language Models
Propose MLingualFC, evaluating vision-language model jailbreak vulnerabilities with multilingual flowchart images
arXiv
2606.07706
Published
Jun 5
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
Qwen2.5-VL-3B-Instruct, Gemma-4-E4B-it, Pangea-7B
- AttacksJailbreaks
- TechniqueCross-modal payloads
- ComponentVision
Paper analysisRead analysis

AI Safety Daily

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