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

Oct 13 · Tuesday

All papers

Found 33 papers

Filter5

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.

System type：web agent System type：Multi-agent systems System type：Embodied systems and robotics System component：Memory System component：Multimodal Clear all

Another 215papers do not have category tags yet and are not included in filtered results.

Risky behaviorarXiv 2610.12822Oct 9
Reward hacking as a transmissible strategy in multi-agent coding systems
Show that reward hacking can propagate through code, memory, and plans in multi-agent coding systems
arXiv
2610.12822
Published
Oct 9
Layer
Application layer
Setting
Multi-agent systems, coding agent
Evaluated models
Qwen2.5-Coder-14B, Gemma-3-12B, Qwen2.5-Coder-7B and others16
- RiskReward hacking
- ComponentMemory
AbstractThe authors state that reward hacking can spread across models and affect routing and aggregation, and that review can reduce delivery.
Reward hacking induced by coding RL can leave the source agent and enter subsequent collaboration and system coordination.
Paper analysisRead analysis
AttacksarXiv 2610.09708Oct 7
研究者提出利用祖先 VLM 的黑盒对抗补丁攻击，可降低 VLA 任务成功率
提出利用祖先 VLM 优化对抗补丁，使已部署 VLA 任务失败
arXiv
2610.09708
Published
Oct 7
Layer
Prompt layer
Setting
Embodied systems and robotics
Evaluated models
OpenVLA, UniVLA, π0.5
- TechniqueGradient and representation optimization
- ComponentVision
Paper analysisRead analysis
Analysis and theoryarXiv 2610.09517Oct 7
审计发现 CLIP 安全分数主要反映画面与提示词场景的相似度
开环审计冻结视觉语言安全分，发现其反映场景模板而非危险感知
arXiv
2610.09517
Published
Oct 7
Layer
Model layer
Setting
Embodied systems and robotics
Evaluated models
CLIP ViT-B/32, CLIP ViT-L/14, Qwen2-VL-7B
- ComponentVision
Abstract论文审计了冻结 VLM 安全评分机制，指出其反映场景模板而非危险感知，为安全强化学习评估提供了严谨基准。
本文是一项针对安全强化学习中冻结视觉语言模型（VLM）安全评分机制的实证审计研究。
Paper analysisRead analysis
DefensesarXiv 2610.09469Oct 7
Secure-CUA: Constraining the Impact of Untrusted Content in Computer-Use Agents
提出 Secure-CUA，用动作事务约束计算机使用智能体的不可信影响
arXiv
2610.09469
Published
Oct 7
Layer
Application layer
Setting
web agent
Evaluated models
Claude Opus 5, Gemini 3.8 Flash, GPT-5.6-Sol
- DefensesInstruction–data separation
- AttacksPrompt injection
- ComponentVision
AbstractSecure-CUA 用每步事务约束不可信影响，良性 TSR 接近 Vanilla-CUA。
Secure-CUA 要限制图形界面里不可信内容对 CUA 决策和 GUI 执行的影响，同时保留合法任务所需的不可信信息。
Paper analysisRead analysis
AttacksarXiv 2610.09240Oct 7
WebMirage hijacks visual web agents' browser operations with adversarial images
提出 WebMirage，用局部对抗图像劫持视觉网页 Agent 的浏览器操作
arXiv
2610.09240
Published
Oct 7
Layer
Application layer
Setting
web agent
Evaluated models
LLaVA-v1.5-13B, LLaVA-v1.6-34B, MiniCPM-o-8B and others6
- AttacksPrompt injection
- TechniqueGradient and representation optimization
- ComponentVision
Abstract论文提出了针对网络智能体的端到端视觉红队框架 WebMirage，揭示了从视觉定位到浏览器执行的安全脆弱性。
WebMirage 是一个面向视觉定位网络智能体的流水线感知端到端红队攻击框架。
Paper analysisRead analysis
DefensesarXiv 2610.07657Oct 6
DEFER1 多智能体防御研究报告受控测试中的攻击成功率下降
实现 DEFER，以确定性检查优先拦截多智能体流水线的内容攻击
arXiv
2610.07657
Published
Oct 6
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
Qwen3-235B-A22B-Instruct-2507, gpt-oss-120b, Llama-4-Scout-17B-16E-Instruct and others5
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentGuardrails and judges
AbstractDEFER 把确定性检查放在 judge 之前。
DEFER 用来测量多智能体授权流水线里，确定性规则在何处结束、LLM judge 从何处开始。
Paper analysisRead analysis
AttacksarXiv 2610.06814Oct 6
TAPDreamer：可跨任务迁移的世界动作模型对抗补丁
TAPDreamer 用公开编码器做固定补丁。
arXiv
2610.06814
Published
Oct 6
Layer
Prompt layer
Setting
Embodied systems and robotics
Evaluated models
FastWAM, DreamWAM-uncond, DreamWAM-joint and others4
- TechniqueGradient and representation optimization
- ComponentVision
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
DefensesarXiv 2610.04616Oct 3
PerturBot 用扰动训练打破视觉-语言-动作模型的模态捷径
提出 Perturbot 扰动训练，打破 VLA 的模态捷径
arXiv
2610.04616
Published
Oct 3
Layer
Training and data layer
Setting
Embodied systems and robotics
Evaluated models
π0.5
- DefensesModel hardening
- ComponentVision
AbstractPerturbot 改训练数据以削弱三类模态捷径，GroundFscore 用来看成功是否来自任务证据。
Perturbot 是一套不改推理的 VLA 训练期数据干预，用来削弱模态捷径，并用离线分数区分“任务变好”和“靠捷径变好”。
Paper analysisRead analysis
DefensesarXiv 2610.03498Oct 2
检测与抑制：针对 VLA 模型对抗补丁的机制可解释性防御
提出条件特征抑制防御，检出对抗补丁后抑制 VLA 内部特征
arXiv
2610.03498
Published
Oct 2
Layer
Model layer
Setting
Embodied systems and robotics
Evaluated models
π0.5, SmolVLA
- DefensesInference-time intervention
- ComponentVision
Abstract条件抑制与 patch 相关的 SAE 特征，间歇攻击下可提高 SR，持续攻击下则不能。
Paper analysisRead analysis
OtherarXiv 2610.03036Oct 2
WebFovea 提出视觉网页智能体四阶段加固框架，WebRetriever 挑战赛得分从 31.0 升至 57.0
提出 WebFovea，加固视觉网页智能体的解析、执行、反馈与观察往返
arXiv
2610.03036
Published
Oct 2
Layer
Application layer
Setting
web agent
Evaluated models
claude-opus-4-6
- ComponentVision
Paper analysisRead analysis
Risky behaviorarXiv 2609.38971Sep 30
研究测量并预测具身 VLM 规划器中任务的拒答可变性
测量并预测具身 VLM 规划器的任务拒答可变性
arXiv
2609.38971
Published
Sep 30
Layer
Model layer
Setting
Embodied systems and robotics
Evaluated models
gemini-robotics-er-2-preview, gemini-robotics-er-1.6-preview
- RiskInappropriate refusal
- ComponentVision
Abstract日常物体能翻转一部分已给出的拒答，可翻转程度取决于任务且可被本地代理预测。
作者测量 constitution 守护的具身 VLM 规划器：一次拒答在只往场景里放一个日常物体之后还保不保持得住，并把这种易翻转程度称为任务的 malleability。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.38948Sep 30
DrivingBench 发布：让视觉语言模型驾驶丰田 Corolla 绕桩
提出 DrivingBench，评测通用视觉语言模型能否闭环驾驶真车
arXiv
2609.38948
Published
Sep 30
Layer
Application layer
Setting
Embodied systems and robotics
Evaluated models
GPT-6 Astra, Claude Fable 5.1, Grok 4.6 and others4
- ComponentVision
Abstract仅 Astra 第二次跑完低速锥桶路线。
DrivingBench 评测通用 VLM 能否自己闭环开真车，推理延迟也算进任务。
Paper analysisRead analysis
AttacksarXiv 2609.38270Sep 29
VirusCascade：劫持 LLM 推荐智能体的协同反思机制
提出 VirusCascade，联合操纵商品描述与交互轨迹以劫持推荐智能体协同反思
arXiv
2609.38270
Published
Sep 29
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
AgentCF, AgentSEQ, AgentRAG and others4
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentMemory
Abstract作者把反思写回视为推广攻击入口，并用联合操纵和诊断探针研究其持续与传播。
VirusCascade 是针对推荐智能体协作反思的 black-box targeted promotion 攻击研究。 攻击者只改商品描述和少量账号的普通交互，试图让目标商品获得更多良性用户曝光。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.36739Sep 29
Frontier Autolab：多智能体 LLM 组织在五十年技术变迁中的长期测试台
评测多智能体 LLM 组织跨时代再创业及历史打分中的事后泄漏
arXiv
2609.36739
Published
Sep 29
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
Claude Opus 5.5, Luna 6
- ComponentMemory
Abstract组织看见转变却建在旁边。
Paper analysisRead analysis
DefensesarXiv 2609.17648Sep 15
多智能体 LLM 流水线中的信任传播与结构隔离
用任务绑定令牌和策略预言机拦住多智能体流水线的未授权执行
arXiv
2609.17648
Published
Sep 15
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
gemma4:31b-cloud
- DefensesPermissions and approval
- AttacksPoisoning and backdoors
- ComponentPermissions and sandboxing
Abstract结构性授权拦住了记忆投毒的未授权执行，持密钥时承重的是策略预言机。
作者在一条会改写沙箱邮件库的四智能体流水线上，检查低权限失陷会不会变成未授权的真实动作。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.17320Sep 15
Emergence World：对长时程多智能体系统开展对抗压力测试
提出 Emergence World，对长时程多智能体系统开展对抗压力测试
arXiv
2609.17320
Published
Sep 15
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
Claude Opus 4.8, DeepSeek v4 Pro, Gemini 3.5 Flash and others9
- AttacksPrompt injection
- RiskMisalignment and value divergence
- ComponentMemory
Abstract作者通过长期平行世界实验指出模型对齐不具备可组合性，单模型群体会产生社会奉承、协同退缩及语言不透明等失效。
Paper analysisRead analysis
OtherarXiv 2609.10144Sep 9
AIOS 的内核管理共享内存：面向系统级个性化的新抽象
提出内核管理共享记忆抽象，统一多智能体的检索、隐私过滤与注入
arXiv
2609.10144
Published
Sep 9
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
GPT-4o, Llama-3.1:8B, Qwen-2.5:7B
- ComponentMemory
Abstract作者称内核集中管理记忆，能拿到无约束上下文的大部分个性化收益，并降低延迟。
kernel-managed shared memory 把多智能体的检索、隐私和注入收进 agent-system kernel，并在 AIOS 上与三种替代方案比较。
Paper analysisRead analysis
AttacksarXiv 2608.10393Aug 11
DURA：用扩散模型生成自然对抗补丁攻击 VLA 机器人模型
提出 DURA，用扩散模型生成自然对抗补丁操控 VLA 机器人动作
arXiv
2608.10393
Published
Aug 11
Layer
Prompt layer
Setting
Embodied systems and robotics
Evaluated models
OpenVLA-7B, π0-FAST, openvla-7B
- TechniqueCross-modal payloads
- ComponentVision
AbstractDURA 用扩散潜空间轨迹生成自然补丁，白盒与黑盒都能把 VLA 导向指定动作。
DURA 是针对 VLA 的扩散式、补丁内容不受限的机器人攻击，同时支持白盒梯度和只看动作输出的黑盒。
Paper analysisRead analysis
Risky behaviorarXiv 2608.10218Aug 11
研究者用进化算法构造多智能体 LLM 系统中的自我传播“思维病毒”
用演化算法构造 mind virus 并测量其在多智能体中的传播
arXiv
2608.10218
Published
Aug 11
Layer
Application layer
Setting
Multi-agent systems, coding agent
Evaluated models
Claude Haiku 4.5, Claude Sonnet 4.6, Gemini 3 Flash and others7
- RiskMisalignment and value divergence
- ComponentMemory
Paper analysisRead analysis

AI Safety Daily

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

All papers

Attack type

Attack technique

Model risks

Impact

System type

Attack surface / entry point

Layer

System component

Reward hacking as a transmissible strategy in multi-agent coding systems

研究者提出利用祖先 VLM 的黑盒对抗补丁攻击，可降低 VLA 任务成功率

审计发现 CLIP 安全分数主要反映画面与提示词场景的相似度

Secure-CUA: Constraining the Impact of Untrusted Content in Computer-Use Agents

WebMirage hijacks visual web agents' browser operations with adversarial images

DEFER1 多智能体防御研究报告受控测试中的攻击成功率下降

TAPDreamer：可跨任务迁移的世界动作模型对抗补丁

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

PerturBot 用扰动训练打破视觉-语言-动作模型的模态捷径

检测与抑制：针对 VLA 模型对抗补丁的机制可解释性防御

WebFovea 提出视觉网页智能体四阶段加固框架，WebRetriever 挑战赛得分从 31.0 升至 57.0

研究测量并预测具身 VLM 规划器中任务的拒答可变性

DrivingBench 发布：让视觉语言模型驾驶丰田 Corolla 绕桩

VirusCascade：劫持 LLM 推荐智能体的协同反思机制

Frontier Autolab：多智能体 LLM 组织在五十年技术变迁中的长期测试台

多智能体 LLM 流水线中的信任传播与结构隔离

Emergence World：对长时程多智能体系统开展对抗压力测试

AIOS 的内核管理共享内存：面向系统级个性化的新抽象

DURA：用扩散模型生成自然对抗补丁攻击 VLA 机器人模型

研究者用进化算法构造多智能体 LLM 系统中的自我传播“思维病毒”