mlsanityData to 7 Oct 2026

ProxyGuard: Direct Reliability Inference for Randomized Data Release Mechanisms with Shared Targets

arXiv 2608.18643 ↗First posted 19 Aug 2026cs.LG stat.ME stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

Researchers often choose a proxy dataset from many releases, transformations, or seeds. Search can make an invalid release appear adequate, while one adequate release does not establish that its generator is reliable. ProxyGuard controls both errors using prespecified bounded risks and a sealed target set. Named-release mode corrects for multiplicity and certifies specific releases. Direct shared-target mode evaluates independent mechanism draws on a common target, lower-bounds their favorable-score rate, and subtracts a bound on favorable scores contributed by invalid releases. Conditional on the target, release scores are independent, yielding a finite-sample mechanism-reliability guarantee without independent target batches or assumptions on release-level p-value dependence. We show that the mean-only penalty is sharp and derive a smooth-score certificate with additive target concentration. In a registered three-requirement study, direct mode raises power from 5.6% to 64.2% at reliability 0.95, while named mode remains stronger under high-signal evidence. Prospective audits span full-pipeline Rice--TVAE, which retrains on every draw, and a non-tabular text mechanism.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 14 of its 37 references that are in the map. First posted on arXiv 19 Aug 2026.

Builds on

14 of 37 references.

References · 14

  1. 2019Modeling Tabular data using Conditional GANcited by 695
  2. 2022TabDDPM: Modelling Tabular Data with Diffusion Modelscited by 373
  3. 2021Learn then Test: Calibrating Predictive Algorithms to Achieve Risk Controlcited by 256
  4. 2021How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Modelscited by 193
  5. 2023Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Spacecited by 183
  6. 2022AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Datacited by 88
  7. 2023Membership Inference Attacks against Synthetic Data through Overfitting Detectioncited by 66
  8. 2023Synthetic data, real errors: how (not) to publish and use synthetic datacited by 29
  9. 2024SynthEval: A Framework for Detailed Utility and Privacy Evaluation of Tabular Synthetic Datacited by 27
  10. 2024Adaptive Learn-then-Test: Statistically Valid and Efficient Hyperparameter Selectioncited by 13
  11. 2023The Real Deal Behind the Artificial Appeal: Inferential Utility of Tabular Synthetic Datacited by 5
  12. 2025Privacy Auditing Synthetic Data Release through Local Likelihood Attackscited by 5
2 more
  1. 2025MIA-EPT: Membership Inference Attack via Error Prediction for Tabular Datacited by 4
  2. 2022Powerful Partial Conjunction Hypothesis Testing via Conditioningcited by 1

Built on by

Not cited yet.