AttacksarXiv 2610.09708
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
- Layer
- Prompt layer
- Evaluated models
- OpenVLA, UniVLA, π0.5
- TechniqueGradient and representation optimization
- ComponentVision
All papers
Filter2
Attack type
Attack technique
Impact
System type
Attack surface / entry point
Layer
System component
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.
Another 233papers do not have category tags yet and are not included in filtered results.
- AttacksarXiv 2610.06814
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
- Layer
- Prompt layer
- Evaluated models
- FastWAM, DreamWAM-uncond, DreamWAM-joint and others4
- TechniqueGradient and representation optimization
- ComponentVision
- AttacksarXiv 2610.06083
研究者提出面向深度强化学习的轨迹级可迁移黑盒对抗攻击
提出轨迹级可迁移对抗攻击，降低深度强化学习策略的期望回报
- arXiv
- 2610.06083
- Published
- Layer
- Model layer
- Evaluated models
- DQN, Double DQN (DDQN)
Abstract轨迹级后退视界攻击在 CartPole 上比逐步 FGSM 类和 Sign 噪声更压低回报。
这项工作研究 DRL 上的迁移式黑盒观测攻击：受害参数不可访问时，在白盒代理上造扰动再迁移。
- AttacksarXiv 2609.22711
UBA-ORL：针对离线强化学习的遗忘激活后门攻击
提出 UBA-ORL，以 BD/CM 样本在遗忘后激活离线 RL 后门
- arXiv
- 2609.22711
- Published
- Evaluated models
- TD3+BC, BCQ, IQL
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
AbstractUBA-ORL 用竞争的 BD/CM 样本，使离线 RL 后门在删除伪装轨迹后激活。
UBA-ORL 研究离线强化学习中由合规轨迹删除激活的后门：合法贡献者同时上传共享触发的后门轨迹和伪装轨迹，训练后触发响应被压低，删除伪装批次后再升高。受害系统是使用静态数据集和轨迹级遗忘接口的离线 RL 服务。对手不能看到训练算法、参数、梯度或具体遗忘算法，但能提交可区分批次，并在部署时向观测注入预定触发。
- AttacksarXiv 2609.07051
TrojanWorld：通过想象引导对世界模型智能体植入后门
提出 TrojanWorld，用物理触发器对世界模型智能体植入想象引导后门
- arXiv
- 2609.07051
- Published
- Evaluated models
- TD-MPC2, DreamerV3, R2-Dreamer
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentFine-tuning and open weights
AbstractTrojanWorld 揭示了世界模型智能体的供应链后门风险，通过在世界模型中操纵想象可实现对闭环动作的持续控制。
- AttacksarXiv 2608.10393
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
- Layer
- Prompt layer
- 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。
- AttacksarXiv 2607.10712
研究提出间接数据投毒攻击，可操纵 Claude、Codex 与 Gemini 的科研结论
提出间接数据投毒攻击，操纵科研智能体得出虚假结论
- arXiv
- 2607.10712
- Published
- Layer
- Application layer
- Setting
- science agent
- Evaluated models
- Claude Code with Claude Opus 4.7, Codex with GPT-5.5, Gemini CLI with Gemini 3.1 Pro
- DefensesInstruction–data separation
- AttacksPoisoning and backdoors
- AttacksarXiv 2606.09499
研究者提出针对世界模型的机器人学习管线投毒攻击
提出 VPH 与 VTH，只改视频帧劫持世界模型并投毒下游策略
- arXiv
- 2606.09499
- Published
- Evaluated models
- Cosmos-Predict 2.5, Cosmos-Predict 2.5 AC, Dino WM and others4
- AttacksPoisoning and backdoors
- TechniqueGradient and representation optimization
- ComponentVideo
Abstract只改视频帧经世界模型投毒。
只改看似安全的视频帧，经世界模型之后才改变下游机器人策略。 问题在于世界模型进入数据生成后，数据集检查看到的仍是安全演示，合成轨迹却可以变危险。
- AttacksarXiv 2604.05595
Diversity-aware red teaming reveals language vulnerabilities in vision-language-action models
Proposes DAERT, which uses simulated failures as reward to generate diverse adversarial instructions, exposing VLA language vulnerabilities.
- arXiv
- 2604.05595
- Published
- Layer
- Prompt layer
- Evaluated models
- π0, OpenVLA, OpenVLA-7B and others5
- TechniqueAutomated search
- AttacksarXiv 2604.03890
From prompts to physical actions: structured backdoor attacks against LLM-mediated robot control systems
Propose a structured JSON backdoor that triggers malicious actions in robot control via a poisoned LoRA
- arXiv
- 2604.03890
- Published
- Evaluated models
- Llama-3.1-8B-Instruct, Gemma-2-9B-IT, DeepSeek-R1-Distill-Llama-8B and others4
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentFine-tuning and open weights
Abstract推理层后门传不到执行。
- AttacksarXiv 2603.24935
SABER: A Stealthy Agentic Black-Box Attack Framework for Vision-Language-Action Models
Propose SABER, which uses a reinforcement learning attacker to rewrite instructions within a budget to cause VLA task failure.
- arXiv
- 2603.24935
- Published
- Layer
- Prompt layer
- Evaluated models
- π0-LIBERO, π0.5, X-VLA and others6
- TechniqueAutomated search
- AttacksarXiv 2603.01414
Jailbreaking embodied LLMs via action-level manipulation
Propose Blindfold, jailbreaking embodied LLMs with action-level manipulation
- arXiv
- 2603.01414
- Published
- Layer
- Application layer
- Attacker
- No-box (offline transfer)
- Evaluated models
- Llama-3.1-8B, DeepSeek-R1-14B, Gemma-3-27B and others8
- AttacksJailbreaks
- TechniqueEncoding and obfuscation
AbstractBlindfold jailbreaks embodied LLMs with action-level proxy planning, and the authors claim it can bypass semantic checks and is effective in both simulation and real-world settings.
Blindfold is a no-box attack against embodied LLM planning modules, rewriting malicious intent into seemingly benign action sequences.
- AttacksarXiv 2601.13612
PINA：针对导航智能体的提示注入攻击
提出 PINA，迭代优化针对导航智能体的提示注入
- arXiv
- 2601.13612
- Published
- Layer
- Application layer
- Evaluated models
- GPT-3.5-turbo, GPT-4, Balcı et al. [2]
- AttacksPrompt injection
- TechniqueAutomated search
AbstractPINA 在黑盒下优化导航提示注入。
PINA 是面向基于 LLM 的导航智能体、在黑盒条件下优化提示注入的框架。
- AttacksarXiv 2505.23266
AdvOF：用对抗物体融合攻击 VLM 驱动的视觉语言导航智能体
提出 AdvOF，生成对抗物体误导视觉语言导航智能体的场景感知
- arXiv
- 2505.23266
- Published
- Layer
- Application layer
- Attacker
- White-box
- Evaluated models
- Lseg, Clip, ViT-L/16 and others5
- AttacksDetection evasion
- TechniqueGradient and representation optimization
- ComponentVision
AbstractAdvOF 以多视角融合生成对抗物体。
AdvOF 是针对 VLM 驱动 VLN 服务导航的对抗物体攻击。
You have reached the end