mlsanityData to 7 Oct 2026

Robustness of on-device Models: Adversarial Attack to Deep Learning Models on Android Apps

arXiv 2101.04401 ↗First posted 12 Jan 2021cs.LG cs.CR cs.SEcited by 19

Not on the map. Why?

Abstract (arXiv)

Deep learning has shown its power in many applications, including object detection in images, natural-language understanding, and speech recognition. To make it more accessible to end users, many deep learning models are now embedded in mobile apps. Compared to offloading deep learning from smartphones to the cloud, performing machine learning on-device can help improve latency, connectivity, and power consumption. However, most deep learning models within Android apps can easily be obtained via mature reverse engineering, while the models' exposure may invite adversarial attacks. In this study, we propose a simple but effective approach to hacking deep learning models using adversarial attacks by identifying highly similar pre-trained models from TensorFlow Hub. All 10 real-world Android apps in the experiment are successfully attacked by our approach. Apart from the feasibility of the model attack, we also carry out an empirical study that investigates the characteristics of deep learning models used by hundreds of Android apps on Google Play. The results show that many of them are similar to each other and widely use fine-tuning techniques to pre-trained models on the Internet.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Builds on 25 of its 45 references that are in the map. First posted on arXiv 12 Jan 2021.

Builds on

25 of 45 references.

References · 25

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  8. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  9. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  10. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,801
  11. 2016Deep Learning with Differential Privacycited by 2,682
  12. 2016Adversarial examples in the physical worldcited by 1,762
13 more
  1. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  2. 2016SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model sizecited by 1,054
  3. 2015Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Bookscited by 1,041
  4. 2016Stealing Machine Learning Models via Prediction APIscited by 602
  5. 2017Recent Trends in Deep Learning Based Natural Language Processingcited by 262
  6. 2018DeepGauge: Multi-Granularity Testing Criteria for Deep Learning Systemscited by 144
  7. 2018Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defensescited by 109
  8. 2019Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Trainingcited by 95
  9. 2018Adversarial Examples: Opportunities and Challengescited by 62
  10. 2018A First Look at Deep Learning Apps on Smartphonescited by 47
  11. 2020Mind Your Weight(s): A Large-scale Study on Insufficient Machine Learning Model Protection in Mobile Appscited by 35
  12. 2018Deep Learning Towards Mobile Applicationscited by 16
  13. 2019Towards Security Threats of Deep Learning Systems: A Surveycited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 2; 2023: 5; 2024: 1; 2025: 5; 2026: 1.

19 papers.

Sort:

By area:

  1. 2021DeepPayload: Black-box Backdoor Attack on Deep Learning Models through Neural Payload Injectioncited by 42
  2. 2021NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identificationcited by 42
  3. 2021A Survey on Data-driven Software Vulnerability Assessment and Prioritizationcited by 18
  4. 2022Smart App Attack: Hacking Deep Learning Models in Android Appscited by 12
  5. 2023Energy-Latency Attacks to On-Device Neural Networks via Sponge Poisoningcited by 10
  6. 2021Piracy-Resistant DNN Watermarking by Block-Wise Image Transformation with Secret Keycited by 9
  7. 2023Pitfalls in Language Models for Code Intelligence: A Taxonomy and Surveycited by 9
  8. 2023Beyond the Model: Data Pre-processing Attack to Deep Learning Models in Android Appscited by 2
  9. 2024A Picture is Worth 500 Labels: A Case Study of Demographic Disparities in Local Machine Learning Models for Instagram and TikTokcited by 2
  10. 2025THEMIS: Towards Practical Intellectual Property Protection for Post-Deployment On-Device Deep Learning Modelscited by 2
  11. 2021Robustness Analysis of Deep Learning Frameworks on Mobile Platformscited by 1
  12. 2022An Empirical Study of AI Techniques in Mobile Applicationscited by 1
7 more
  1. 2023Pairwise GUI Dataset Construction Between Android Phones and Tabletscited by 1
  2. 2025Stealthy Backdoor Attack to Real-world Models in Android Appscited by 1
  3. 2025"Impressively Scary:" Exploring User Perceptions and Reactions to Unraveling Machine Learning Models in Social Media Applicationscited by 1
  4. 2023Towards Real Smart Apps: Investigating Human-AI Interactions in Smartphone On-Device AI Appscited by 0
  5. 2025NeuroDeX: Unlocking Diverse Support in Decompiling Deep Neural Network Executablescited by 0
  6. 2025BinCtx: Multi-Modal Representation Learning for Robust Android App Behavior Detectioncited by 0
  7. 2026SoK: Attack and Defense Landscape of Mobile On-device AI Systemscited by 0