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.

System component：Permissions and sandboxing System component：Video Clear all

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

Evaluation and benchmarksarXiv 2610.07939Oct 6
CCDF dataset: Grok Imagine, Veo 3.1, Sora 2 forge surveillance footage, ten deepfake detectors basically cannot recognize it
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
Evaluation and benchmarksarXiv 2610.07639Oct 6
HarnessSecurity-Bench horizontally evaluates the security mechanisms of six coding agent frameworks.
评测编码智能体 harness 安全机制在对抗任务中的防护效果
arXiv
2610.07639
Published
Oct 6
Layer
Application layer
Setting
coding agent
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.07359Oct 6
论文实测 Agent 工具调用门控栈的失败相关性
提出 nmult 并评测 Agent 动作门禁堆叠的错误相关性
arXiv
2610.07359
Published
Oct 6
Layer
Application layer
Setting
AI Agent
Evaluated models
AgentTrust runtime (174 rules), GPT-5.5 (gpt55), claude-haiku-4-5-20251001 and others7
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract论文实测发现大模型裁判间错误显著相关，作者主张按机制选型并在整栈层面度量防护增益。
论文是一项针对智能体动作拦截防御堆叠故障关联性与实际组合价值的实证测量研究。
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
Evaluation and benchmarksarXiv 2610.06171Oct 5
ControlPed 生成逼真行人危险动作场景，七款端到端驾驶模型 HDScore 从 88.8 降至 47.4
提出 ControlPed 与 HazardPed，生成可控行人危险场景并评测端到端驾驶
arXiv
2610.06171
Published
Oct 5
Layer
Model layer
Setting
Embodied systems and robotics
Evaluated models
DiffusionDrive, SSR, VAD and others7
- ComponentVideo
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.05840Oct 5
研究：任务范围授权可将工具调用有害执行降至零
用配对回放评测任务范围授权能否阻断越权工具执行
arXiv
2610.05840
Published
Oct 5
Layer
Application layer
Setting
AI Agent
Evaluated models
Llama 3.2 3B Instruct Q4 K M, Qwen3 4B Instruct, Gemma 3 4B and others8
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.02999Oct 2
OmniConfess：用 Token 级「忏悔」缓解全模态幻觉
提出 OmniConfess，冻结候选并按通道重打分以缓解全模态幻觉
arXiv
2610.02999
Published
Oct 2
Layer
Model layer
Setting
Models and APIs
Evaluated models
Qwen2.5-Omni-7B, Qwen3-Omni-30B-A3B, Nemotron-3-Nano-Omni-30B
- DefensesInference-time intervention
- RiskHallucination and fabrication
- ComponentVision
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.02664Oct 2
论文提出 GHOST：长程智能体在良性对话中遗忘早期安全约束
提出 STAR-Guard 双层防御，抑制长程 Agent 遗忘跨轮安全约束
arXiv
2610.02664
Published
Oct 2
Layer
Application layer
Setting
AI Agent
Evaluated models
GPT-5.5, DeepSeek-V4-Pro, GLM-5.2 and others7
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract揭示 GHOST 跨轮约束遗忘风险，构建 SCARBench 基准并通过 STAR-Guard 双层防御消除违规。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.00392Sep 30
A2A-TIBA：针对 A2A 协议 LLM 智能体的间接提示注入攻击与三层同构攻防模型
提出 A2A-TIBA，经 A2A 对等任务间接注入并植入常驻回调
arXiv
2610.00392
Published
Sep 30
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
MiniMax-M3, deepseek-v4-pro, glm-5.3 and others8
- AttacksPrompt injection
- TechniqueMulti-turn escalation
- ComponentMCP and skills
AbstractA2A-TIBA 一次植入后绕过智能体，GDA 用四类结果定位防线，ELA-ITL 把信封标注列为第三层。
A2A-TIBA 把间接提示注入和绕过放进 A2A：gray-box 攻击者已能投递任务，诱导目标部署常驻回调程序后，下令和外传不再经过智能体。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.35557Sep 29
论文：容器、权限规则与沙箱都无法区分读取研究代码的是编译器还是编码 Agent
评测容器、权限规则与沙箱能否区分编译器与编码 Agent 的文件读取
arXiv
2609.35557
Published
Sep 29
Layer
Application layer
Setting
coding agent
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract此 harness 表达不了编译器可读而智能体不可读，五条路径两边都不覆盖。
作者在物理流求解器上检查一套编码智能体配置：编译器必须读本组材料与反应模型，这些字节又不得进入模型上下文或离开本机的记录。
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
Evaluation and benchmarksarXiv 2609.26900Sep 23
Ajar：在 AgentDojo 上测量 Agent 防御中未关闭的权限
提出 Ajar，测量 Agent 防御对任务不需要的工具调用的放行
arXiv
2609.26900
Published
Sep 23
Layer
Application layer
Setting
AI Agent
Evaluated models
Sonnet-5, Haiku-4.5
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
AbstractAjar 用宿主基准的真值给防御的开放权限打分，OPL 独立于攻击成功和效用。
Ajar 量的是智能体防御留下的开放权限：任务不需要、但防御仍会允许的工具调用。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.22076Sep 19
APort Vault released: evaluating AI agent payment authorization with 4,371 human-written attacks
构建 APort Vault 基准，对比支付智能体模型拒答与确定性授权
arXiv
2609.22076
Published
Sep 19
Layer
Application layer
Setting
AI Agent
Evaluated models
Claude Fable 5.1, Claude Sonnet 5, Claude Haiku 4.5 and others14
- DefensesPermissions and approval
- AttacksJailbreaks
- ComponentPermissions and sandboxing
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2608.18607Aug 19
VA-Judger：面向联合视频-音频生成的人类偏好反馈奖励建模
提出 VA-Judger，用人类偏好训练联合音视频奖励模型
arXiv
2608.18607
Published
Aug 19
Layer
Training and data layer
Setting
Models and APIs
Evaluated models
VA-Judger, LTX-2 + VA-Judger
- DefensesModel hardening
- RiskReward hacking
- ComponentGuardrails and judges
AbstractVA-Judger 用易到难的人类偏好学习做联合音视频奖励，并用于 LTX-2 后训练。
VA-Judger 是面向联合音视频生成的思维链全模态奖励模型，用来替代由分维度专家指标拼成的强化学习奖励。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2608.09476Aug 10
ActBench：面向协作智能体行为安全的自演化基准
构建自演化基准 ActBench，评测协作智能体的行为安全
arXiv
2608.09476
Published
Aug 10
Layer
Application layer
Setting
AI Agent
Evaluated models
Claude-Opus-4.8, Claude-Sonnet-4.6, GPT-5.5 and others15
- AttacksPrompt injection
- ComponentPermissions and sandboxing
AbstractActBench 通过自演化搜索与双证据重构评测轨迹行为安全，揭示基础模型决策差异主导了协同智能体的操作风险。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2608.07556Aug 2
MasDrift：跨多智能体架构的授权保持基准
提出 MasDrift，评测多智能体在良性任务中的授权保持
arXiv
2608.07556
Published
Aug 2
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
DeepSeek V4 Flash, Qwen3.7 Plus, GPT-5.4 Nano and others6
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract授权保持可被测量：层级用未授权动作换完成率，重锚定比沿链携带更稳。
MasDrift 把多智能体设计中的授权保持做成可测量性质，对象是没有攻击者的普通委托。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2605.18984May 19
Artifact-Bench：评估 MLLM 检测与诊断 AI 生成视频伪影
提出 Artifact-Bench，评测 MLLM 检测并诊断 AI 生成视频伪影
arXiv
2605.18984
Published
May 19
Layer
Model layer
Setting
Models and APIs
Evaluated models
Gemini 3.1 Pro, Gemini 3 Flash, Qwen3-VL 8B-Instruct and others19
- DefensesDetection and filtering
- ComponentVideo
AbstractArtifact-Bench 以三项任务评 MLLM 的 AIGC 伪影感知。
Artifact-Bench 用来看 MLLM 能否发现 AI 生成视频中的伪影，并说明视频为何不真实。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2605.14859May 14
论文提出 AuthBench 基准，评测编码 Agent 能否推断最小权限边界
提出 AuthBench 评测编码 Agent 最小权限，并验证充足性-紧缩性分解
arXiv
2605.14859
Published
May 14
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-5, GPT-5.3-Codex, GPT-5.4 and others9
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract论文提出权限边界推断基准与两阶段分解方法，有效降低安全暴露。
本文研究代码智能体在执行终端任务前自主推断最小文件权限边界的能力。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2602.10639Feb 11
VideoSTF 揭示视频大模型输出重复问题
提出 VIDEO STF 评测视频大模型输出重复并压力测试时序扰动
arXiv
2602.10639
Published
Feb 11
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
LLaVA-Video-7B-Qwen2, LLaVA-Video-7B-Qwen2-Video-Only, LLaVA-NeXT-Video-7B and others10
- AttacksDenial of service
- ComponentVideo
AbstractVIDEO STF 把输出重复做成可测、可被时间扰动放大、并可黑盒诱导的失败。
VIDEO STF 测量 VideoLLM 的输出重复，并把时间变换用作 black-box 诱导。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2512.03992Dec 4, 2025
DIQ-H 基准与 VGIR 框架：评估 VLM 在对抗视觉条件下的鲁棒性
提出 DIQ-H 基准，评测 VLM 在连续退化视觉流中的幻觉残留与恢复
arXiv
2512.03992
Published
Dec 4, 2025
Layer
Model layer
Setting
Models and APIs
Evaluated models
LLaVA-7B-strict, InternVL2-8B-strict, Phi-3-Vision-strict and others17
- RiskHallucination and fabrication
- ComponentVideo
AbstractDIQ-H 用连续退化视频测幻觉延续与恢复，VIR 用不确定性过滤生成伪标注。
DIQ-H 评测 VLM 在连续视觉退化中的幻觉、恢复和跨时间一致性。VIR 用轻量模型的不确定性过滤生成伪真值。
Paper analysisRead analysis

You have reached the end

AI Safety Daily

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