Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 203papers have no tags yet and are not included in filtered results.

System type：coding agent System type：Browser and computer use System type：Multi-agent systems System type：Embodied systems and robotics System component：Memory System component：Multimodal Clear all

Another 203papers 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
DefensesarXiv 2610.11942Oct 8
LGWM: Verifying GUI agent screen transitions with action-conditioned world models
Proposes LGWM, using an action-conditioned world model to verify GUI agent screen transitions.
arXiv
2610.11942
Published
Oct 8
Layer
Application layer
Setting
Browser and computer use
Evaluated models
LGWM
- DefensesMonitoring and auditing
- AttacksPrompt injection
- ComponentMonitors
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
Evaluation and benchmarksarXiv 2610.08902Oct 7
论文提出 agent plasticity 指标衡量智能体通过经验自我改进的效率
提出 agent plasticity 指标衡量智能体通过经验自我改进的效率
arXiv
2610.08902
Published
Oct 7
Layer
Application layer
Setting
coding agent
Evaluated models
Claude Fable 5, Claude Opus 5, GPT-5.6 Sol and others10
- ComponentMemory
Abstract作者用固定权重协议比较前沿模型把经验写成制品后的留出增益与习得效率。
作者用 agent plasticity 衡量一件事：权重冻结时，智能体把经验写成可继承制品，再换成留出能力增益，这个转换有多省。
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 2610.05241Oct 4
StateWise: Diagnosing and repairing persistent operation state before agent actions
Propose StateWise, which diagnoses and repairs invalid persistent state before coding agent actions
arXiv
2610.05241
Published
Oct 4
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-5.6-sol, DeepSeek v4 Pro, Qwen 3.7 Max and others4
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentMemory
AbstractStateWise repairs stale persistent premises before action.
StateWise is a framework that diagnoses and repairs the persistent operational state of coding agents before executing actions with side effects.
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
DefensesarXiv 2610.00450Oct 1
ZoneClaw 用记忆分区防御 OpenClaw 式计算机使用 Agent 的持久记忆攻击
提出 ZoneClaw，用记忆分区防御计算机使用 Agent 的持久记忆攻击
arXiv
2610.00450
Published
Oct 1
Layer
Application layer
Setting
Browser and computer use
Evaluated models
Claude Sonnet 4.6, GPT-5.5, GLM-5.2 and others4
- DefensesInstruction–data separation
- AttacksPoisoning and backdoors
- ComponentMemory
Paper analysisRead analysis
Risky behaviorarXiv 2609.38971Sep 30
The study measures and predicts refusal variability across tasks in embodied VLM planners
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: A Long-Horizon Testbed for Multi-Agent LLM Organizations Across Fifty Years of Technological Change
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
Evaluation and benchmarksarXiv 2610.06898Sep 29
DIBench：面向移动 GUI 智能体多候选选择决策完整性的安全基准
用 DIBench 评测移动 GUI 智能体被 UI 注入后的决策完整性
arXiv
2610.06898
Published
Sep 29
Layer
Application layer
Setting
Browser and computer use
Evaluated models
Qwen2.5-VL(32B), Qwen3-VL(32B), GUI-Owl(32B) and others7
- AttacksPrompt injection
- ComponentVision
Abstract论文提出移动智能体决策完整性基准 DIBench，发现欺诈 UI 注入诱导目标选择并推高完成率，通用防御收益不稳定。
Paper analysisRead analysis
Risky behaviorarXiv 2609.35291Sep 28
窄域多模态微调可诱发涌现性失配，覆盖 15 个视觉语言模型
证明无显式危害的图文窄微调可诱发跨任务涌现未对齐
arXiv
2609.35291
Published
Sep 28
Layer
Model layer
Setting
Browser and computer use
Evaluated models
GPT-4o, GPT-4.1, Gemini 2.5 and others15
- RiskMisalignment and value divergence
- ComponentVision
Abstract证实窄多模态微调可引发全面的多渠道未对齐行为转变，揭示了后训练阶段对齐被隐蔽重塑的安全隐患。
论文系统探究了视觉-语言模型中由窄任务微调诱发的跨模态涌现未对齐（Emergent Misalignment, EM）现象。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.35912Sep 28
MMSkillRisk：多模态技能图像中的隐藏指令可诱导 Agent 执行未授权操作
提出 MMSkillRisk 基准，评测多模态技能图像中的隐藏指令诱导 Agent 越权操作
arXiv
2609.35912
Published
Sep 28
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-5.6-sol, Kimi-K2.6, Kimi-K3 and others6
- AttacksPrompt injection
- TechniqueCross-modal payloads
- ComponentMCP and skills
Abstract论文提出 MMSkillRisk 基准与 NCVA 攻击，揭示多模态技能中教学图像潜藏指令可引发未授权操作并与良性任务共存。
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

AI Safety Daily

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