mlsanityData to 7 Oct 2026

Manipulation Facing Threats: Evaluating Physical Vulnerabilities in End-to-End Vision Language Action Models

arXiv 2409.13174 ↗First posted 20 Sep 2024cs.CVcited by 16

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

also in 1 more
Abstract (arXiv)

Recently, driven by advancements in Multimodal Large Language Models (MLLMs), Vision Language Action Models (VLAMs) are being proposed to achieve better performance in open-vocabulary scenarios for robotic manipulation tasks. Since manipulation tasks involve direct interaction with the physical world, ensuring robustness and safety during the execution of this task is always a very critical issue. In this paper, by synthesizing current safety research on MLLMs and the specific application scenarios of the manipulation task in the physical world, we comprehensively evaluate VLAMs in the face of potential physical threats. Specifically, we propose the Physical Vulnerability Evaluating Pipeline (PVEP) that can incorporate as many visual modal physical threats as possible for evaluating the physical robustness of VLAMs. The physical threats in PVEP specifically include Out-of-Distribution, Typography-based Visual Prompt, and Adversarial Patch Attacks. By comparing the performance fluctuations of VLAMs before and after being attacked, we provide generalizable textbftextitAnalyses of how VLAMs respond to different physical threats.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 47 of its 60 references that are in the map. First posted on arXiv 20 Sep 2024.

Builds on

47 of 60 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024From LLMs to MLLMs: Exploring the Landscape of Multimodal Jailbreaking21 Jun
  2. 2024An Image Is Worth 1000 Lies: Adversarial Transferability across Prompts on Vision-Language Models14 Mar
  3. 2023Defense-Prefix for Preventing Typographic Attacks on CLIP10 Apr

Other references · 44

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2023Visual Instruction Tuningcited by 7,792
  5. 2017Mask R-CNNcited by 7,259
  6. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  7. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  8. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,337
  9. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  10. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  11. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  12. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,037
32 more
  1. 2023Jailbroken: How Does LLM Safety Training Fail?cited by 1,565
  2. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  3. 2023PaLM 2 Technical Reportcited by 1,154
  4. 2023A Survey on Multimodal Large Language Modelscited by 898
  5. 2022A-OKVQA: A Benchmark for Visual Question Answering using World Knowledgecited by 702
  6. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 678
  7. 2024Evaluating Real-World Robot Manipulation Policies in Simulationcited by 484
  8. 2021Implicit Behavioral Cloningcited by 442
  9. 2023Catastrophic Jailbreak of Open-source LLMs via Exploiting Generationcited by 439
  10. 2022VIMA: General Robot Manipulation with Multimodal Promptscited by 403
  11. 2023A Survey of Hallucination in Large Foundation Modelscited by 365
  12. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  13. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 329
  14. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 276
  15. 2024A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Modelscited by 266
  16. 2023Certifying LLM Safety against Adversarial Promptingcited by 248
  17. 2023TidyBot: Personalized Robot Assistance with Large Language Modelscited by 219
  18. 2023LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examplescited by 177
  19. 2018TallyQA: Answering Complex Counting Questionscited by 152
  20. 2021Towards Robust Vision Transformercited by 137
  21. 2014A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Inputcited by 112
  22. 2024Hierarchical Diffusion Policy for Kinematics-Aware Multi-Task Robotic Manipulationcited by 92
  23. 2024LLaRA: Supercharging Robot Learning Data for Vision-Language Policycited by 51
  24. 2021Triggering Failures: Out-Of-Distribution detection by learning from local adversarial attacks in Semantic Segmentationcited by 40
  25. 2021Pyramid Adversarial Training Improves ViT Performancecited by 39
  26. 2024Alpaca against Vicuna: Using LLMs to Uncover Memorization of LLMscited by 33
  27. 2023Transferring Foundation Models for Generalizable Robotic Manipulationcited by 18
  28. 2024Diffusion Policy Attacker: Crafting Adversarial Attacks for Diffusion-based Policiescited by 17
  29. 2024MMRo: Are Multimodal LLMs Eligible as the Brain for In-Home Robotics?cited by 14
  30. 2024The New Agronomists: Language Models are Experts in Crop Managementcited by 12
  31. 2022Goal-Aware Generative Adversarial Imitation Learning from Imperfect Demonstration for Robotic Cloth Manipulationcited by 7
  32. 2023Identifying Expert Behavior in Offline Training Datasets Improves Behavioral Cloning of Robotic Manipulation Policiescited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 11; 2026: 5.

16 papers.

Sort:

By area:

  1. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applicationscited by 82
  2. 2025Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challengescited by 72
  3. 2025Safety at Scale: A Comprehensive Survey of Large Model and Agent Safetycited by 63
  4. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 46
  5. 2025BadVLA: Towards Backdoor Attacks on Vision-Language-Action Models via Objective-Decoupled Optimizationcited by 37
  6. 2025Eva-VLA: Evaluating Vision-Language-Action Models' Robustness Under Real-World Physical Variationscited by 24
  7. 2025Jailbreak-AudioBench: In-Depth Evaluation and Analysis of Jailbreak Threats for Large Audio Language Modelscited by 17
  8. 2025When Alignment Fails: Multimodal Adversarial Attacks on Vision-Language-Action Modelscited by 16
  9. 2025Phantom Menace: Exploring and Enhancing the Robustness of VLA Models Against Physical Sensor Attackscited by 9
  10. 2026Inject Once Survive Later: Backdooring Vision-Language-Action Models to Persist Through Downstream Fine-tuningcited by 7
  11. 2025A Comprehensive Survey on Physical Risk Control in the Era of Foundation Model-enabled Roboticscited by 2
  12. 2025ANNIE: Be Careful of Your Robotscited by 2
4 more
  1. 2026SABER: A Stealthy Agentic Black-Box Attack Framework for Vision-Language-Action Modelscited by 2
  2. 2026FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Modelscited by 1
  3. 2026VLAGuard: A Framework for Evaluating and Mitigating Physical Attention Hijacking in Vision-Language-Action Robots within Wireless Sensor Networkscited by 0
  4. 2026Structure-Aware Robust Fine-Tuning: Defending Vision-Language-Action Robots Against Physical Attention Hijackingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics
  2. 2025LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models
  3. 2026Exploiting Vulnerabilities: Universal Adversarial Attacks on Vision-Language-Action Models in Robotics
  4. 2026Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms
  5. 2026Hidden in Plain Sight: Diffusion-Based Unrestricted Robotic Attacks on Vision-Language-Action Models
  6. 2025Adversarial Attacks on Robotic Vision Language Action Models
  7. 2025AttackVLA: Benchmarking Adversarial and Backdoor Attacks on Vision-Language-Action Models
  8. 2026Hijacking Robots with a Piece of Paper: A Systematic Study of Physical Prompt Injection in VLM-Controlled Robots