mlsanityData to 14 Oct 2026

diffGHOST: Diffusion based Generative Hedged Oblivious Synthetic Trajectories

arXiv 2605.10647 ↗First posted 11 May 2026cs.AI cs.CRcited by 0

On the map: Diffusion models → Robustness, privacy and adversarial aspects

Abstract (arXiv)

Trajectories are nowadays valuable information for a wide range of applications. However they are also inherently sensitive, as they contain highly personal information about individuals. Facing this challenge, synthesizing mobility trajectories has emerged as a promising solution to leverage mobility information while preserving privacy. State-of-the-art models, often rely on the false assumptions of generative models implicit privacy and fails to provide privacy guarantees while preserving trajectories utility. Here, we introduce diffGHOST, a conditional diffusion model based on latent space segmentation, designed to answer this challenge. Thus, this paper propose a methodology that identify and mitigate memorization of critical samples using condition segments of a learn latent space.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 23 of its 48 references that are in the map. First posted on arXiv 11 May 2026.

Builds on

23 of 48 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023Membership Inference Attacks against Diffusion Models7 Feb
  2. 2023Extracting Training Data from Diffusion Models30 Jan

Other references · 21

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2020Denoising Diffusion Implicit Modelscited by 9,781
  5. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  6. 2022Classifier-Free Diffusion Guidancecited by 5,847
  7. 2022Generative Adversarial Networkscited by 5,627
  8. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  9. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  10. 2025Why Diffusion Models Don't Memorize: The Role of Implicit Dynamical Regularization in Trainingcited by 81
  11. 2023DiffTraj: Generating GPS Trajectory with Diffusion Probabilistic Modelcited by 65
  12. 2020LSTM-TrajGAN: A Deep Learning Approach to Trajectory Privacy Protectioncited by 57
9 more
  1. 2022TAPAS: a Toolbox for Adversarial Privacy Auditing of Synthetic Datacited by 57
  2. 2023Achilles' Heels: Vulnerable Record Identification in Synthetic Data Publishingcited by 28
  3. 2025Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Modelscited by 25
  4. 2023Synthetic is all you need: removing the auxiliary data assumption for membership inference attacks against synthetic datacited by 18
  5. 2024TrajGPT: Controlled Synthetic Trajectory Generation Using a Multitask Transformer-Based Spatiotemporal Modelcited by 13
  6. 2016Individual Differential Privacy: A Utility-Preserving Formulation of Differential Privacy Guaranteescited by 10
  7. 2024SoK: Can Trajectory Generation Combine Privacy and Utility?cited by 6
  8. 2020Smart Urban Mobility: When Mobility Systems Meet Smart Datacited by 3
  9. 2026A Dual Perspective on Synthetic Trajectory Generators: Utility Framework and Privacy Vulnerabilitiescited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Privacy Evaluation of Generative Models for Trajectory Generation
  2. 2025RAPID: Retrieval Augmented Training of Differentially Private Diffusion Models
  3. 2025Leveraging the Spatial Hierarchy: Coarse-to-fine Trajectory Generation via Cascaded Hybrid Diffusion
  4. 2026MobiDiff: Semantic-Aware Multi-Channel Discrete Diffusion for Human Mobility Data Generation
  5. 2026SynHAT: A Two-stage Coarse-to-Fine Diffusion Framework for Synthesizing Human Activity Traces
  6. 2024PateGail: A Privacy-Preserving Mobility Trajectory Generator with Imitation Learning
  7. 2025Latent Diffusion Unlearning: Protecting Against Unauthorized Personalization Through Trajectory Shifted Perturbations
  8. 2026Adaptive Diffusion Freezing: Privacy-preserving Diffusion Models Against Membership Inference Attacks