mlsanityData to 14 Oct 2026

SoK: Attack and Defense Landscape of Mobile On-device AI Systems

arXiv 2607.00362 ↗First posted 1 Jul 2026cs.CR cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Mobile on-device AI (MoAI) systems that integrate locally deployed AI models with conventional mobile software components are emerging as a key paradigm for delivering intelligent functionality directly on end-user devices. By moving inference from remote cloud services to the local mobile environment, such systems enable privacy-preserving, low-latency, and offline-capable AI functionality, yet introduce new security risks arising from the local storage of AI models. This paper presents the first comprehensive systematization of knowledge on MoAI security, covering security pillars, attack landscape, and defense landscape of MoAI systems. We further identify unresolved gaps in current attack and defense research and point to promising directions for future research in this emerging area. Our work establishes the first systematic framework for understanding the attack and defense landscapes of MoAI systems, serving as a foundation for building secure MoAI systems and advancing research in this critical domain. Companion resources are available at https://github.com/Jinxhy/Awesome-MoAI-Security.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 62 references that are in the map. First posted on arXiv 1 Jul 2026.

Builds on

33 of 62 references.

References · 33

  1. 2020DarkneTZ: Towards Model Privacy at the Edge using Trusted Execution Environmentscited by 58
  2. 2018A First Look at Deep Learning Apps on Smartphonescited by 47
  3. 2021DeepPayload: Black-box Backdoor Attack on Deep Learning Models through Neural Payload Injectioncited by 44
  4. 2020Mind Your Weight(s): A Large-scale Study on Insufficient Machine Learning Model Protection in Mobile Appscited by 36
  5. 2021Quantization Backdoors to Deep Learning Commercial Frameworkscited by 29
  6. 2020ShadowNet: A Secure and Efficient On-device Model Inference System for Convolutional Neural Networkscited by 21
  7. 2023No Privacy Left Outside: On the (In-)Security of TEE-Shielded DNN Partition for On-Device MLcited by 21
  8. 2021Robustness of on-device Models: Adversarial Attack to Deep Learning Models on Android Appscited by 19
  9. 2023NNSplitter: An Active Defense Solution for DNN Model via Automated Weight Obfuscationcited by 16
  10. 2022Understanding Real-world Threats to Deep Learning Models in Android Appscited by 15
  11. 2022Smart App Attack: Hacking Deep Learning Models in Android Appscited by 12
  12. 2023Energy-Latency Attacks to On-Device Neural Networks via Sponge Poisoningcited by 10
21 more
  1. 2023MirrorNet: A TEE-Friendly Framework for Secure On-device DNN Inferencecited by 9
  2. 2020Offline Model Guard: Secure and Private ML on Mobile Devicescited by 7
  3. 2023ModelObfuscator: Obfuscating Model Information to Protect Deployed ML-based Systemscited by 7
  4. 2024Investigating White-Box Attacks for On-Device Modelscited by 6
  5. 2025TZ-LLM: Protecting On-Device Large Language Models with Arm TrustZonecited by 5
  6. 2024MalModel: Hiding Malicious Payload in Mobile Deep Learning Models with Black-box Backdoor Attackcited by 4
  7. 2025Efficient Compositional Multi-tasking for On-device Large Language Modelscited by 4
  8. 2023A First Look at On-device Models in iOS Appscited by 3
  9. 2024TinyML Security: Exploring Vulnerabilities in Resource-Constrained Machine Learning Systemscited by 3
  10. 2024TEESlice: Protecting Sensitive Neural Network Models in Trusted Execution Environments When Attackers have Pre-Trained Modelscited by 3
  11. 2021LEAP: TrustZone Based Developer-Friendly TEE for Intelligent Mobile Appscited by 2
  12. 2023Beyond the Model: Data Pre-processing Attack to Deep Learning Models in Android Appscited by 2
  13. 2023Improving End-to-End Neural Diarization Using Conversational Summary Representationscited by 2
  14. 2024Model-less Is the Best Model: Generating Pure Code Implementations to Replace On-Device DL Modelscited by 2
  15. 2024DynaMO: Protecting Mobile DL Models through Coupling Obfuscated DL Operatorscited by 2
  16. 2025THEMIS: Towards Practical Intellectual Property Protection for Post-Deployment On-Device Deep Learning Modelscited by 2
  17. 2024SoK: Towards Security and Safety of Edge AIcited by 1
  18. 2025Stealthy Backdoor Attack to Real-world Models in Android Appscited by 1
  19. 2025TensorShield: Safeguarding On-Device Inference by Shielding Critical DNN Tensors with TEEcited by 1
  20. 2026MirageNet:A Secure, Efficient, and Scalable On-Device Model Protection in Heterogeneous TEE and GPU Systemcited by 1
  21. 2026FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolationcited by 1

Built on by

Not cited yet.