mlsanityData to 14 Oct 2026

A Dual Perspective on Synthetic Trajectory Generators: Utility Framework and Privacy Vulnerabilities

arXiv 2604.19653 ↗First posted 21 Apr 2026cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Human mobility data are used in numerous applications, ranging from public health to urban planning. Human mobility is inherently sensitive, as it can contain information such as religious beliefs and political affiliations. Historically, it has been proposed to modify the information using techniques such as aggregation, obfuscation, or noise addition, to adequately protect privacy and eliminate concerns. As these methods come at a great cost in utility, new methods leveraging development in generative models, were introduced. The extent to which such methods answer the privacy-utility trade-off remains an open problem. In this paper, we introduced a first step towards solving it, by the introduction and application of a new framework for utility evaluation. Furthermore, we provide evidence that privacy evaluation remains a great challenge to consider and that it should be tackled through adversarial evaluation in accordance with the current EU regulation. We propose a new membership inference attack against a subcategory of generative models, even though this subcategory was deemed private due to its resistance over the trajectory user-linking problem.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 35 of its 118 references that are in the map. First posted on arXiv 21 Apr 2026.

Builds on

35 of 118 references.

References · 35

  1. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  2. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  3. 2017Improved Training of Wasserstein GANscited by 2,235
  4. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  5. 2021Membership Inference Attacks From First Principlescited by 878
  6. 2019Modeling Tabular data using Conditional GANcited by 699
  7. 2018ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Modelscited by 498
  8. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 497
  9. 2017Towards Principled Methods for Training Generative Adversarial Networkscited by 464
  10. 2022TabDDPM: Modelling Tabular Data with Diffusion Modelscited by 377
  11. 2017Real-valued (Medical) Time Series Generation with Recurrent Conditional GANscited by 333
  12. 2019Human Motion Trajectory Prediction: A Surveycited by 232
23 more
  1. 2022Reconstructing Training Data with Informed Adversariescited by 148
  2. 2013Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifierscited by 130
  3. 2019Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learningcited by 122
  4. 2021Winning the NIST Contest: A scalable and general approach to differentially private synthetic datacited by 102
  5. 2020A Survey on Deep Learning for Human Mobilitycited by 88
  6. 2021Learning in High Dimension Always Amounts to Extrapolationcited by 66
  7. 2023Membership Inference Attacks against Synthetic Data through Overfitting Detectioncited by 66
  8. 2020LSTM-TrajGAN: A Deep Learning Approach to Trajectory Privacy Protectioncited by 57
  9. 2022TAPAS: a Toolbox for Adversarial Privacy Auditing of Synthetic Datacited by 57
  10. 2024ControlTraj: Controllable Trajectory Generation with Topology-Constrained Diffusion Modelcited by 49
  11. 2022Are Your Sensitive Attributes Private? Novel Model Inversion Attribute Inference Attacks on Classification Modelscited by 43
  12. 2023White-box Membership Inference Attacks against Diffusion Modelscited by 42
  13. 2020OpenStreetMap: Challenges and Opportunities in Machine Learning and Remote Sensingcited by 30
  14. 2023A Linear Reconstruction Approach for Attribute Inference Attacks against Synthetic Datacited by 24
  15. 2023Synthetic is all you need: removing the auxiliary data assumption for membership inference attacks against synthetic datacited by 18
  16. 2018A survey on trajectory clustering analysiscited by 12
  17. 2018The Long Road to Computational Location Privacy: A Surveycited by 10
  18. 2018Generative Models for Simulating Mobility Trajectoriescited by 10
  19. 2025Winning the MIDST Challenge: New Membership Inference Attacks on Diffusion Models for Tabular Data Synthesiscited by 10
  20. 2021Privacy-Preserving Synthetic Location Data in the Real Worldcited by 8
  21. 2024A Zero Auxiliary Knowledge Membership Inference Attack on Aggregate Location Datacited by 4
  22. 2018Mapping the Privacy-Utility Tradeoff in Mobile Phone Data for Developmentcited by 1
  23. 2026Rethinking Anonymity Claims in Synthetic Data Generation: A Model-Centric Privacy Attack Perspectivecited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026diffGHOST: Diffusion based Generative Hedged Oblivious Synthetic Trajectoriescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Privacy Evaluation of Generative Models for Trajectory Generation
  2. 2024PateGail: A Privacy-Preserving Mobility Trajectory Generator with Imitation Learning
  3. 2025What is the Cost of Differential Privacy for Deep Learning-Based Trajectory Generation?
  4. 2025SoK: Can Synthetic Images Replace Real Data? A Survey of Utility and Privacy of Synthetic Image Generation
  5. 2023Privacy and Copyright Protection in Generative AI: A Lifecycle Perspective
  6. 2025Quantifying the Privacy Implications of High-Fidelity Synthetic Network Traffic
  7. 2025When Privacy Isn't Synthetic: Hidden Data Leakage in Generative AI Models
  8. 2026Neighborhood Watch: Privacy Risks in Seeded Local Combination Synthetic Data ↗