SynthGuard-ReleaseBench: Locked-Audit Evidence for Synthetic Tabular Data Releases
Abstract
Synthetic tabular data are often judged by realism, privacy, or downstream-task scores. Those scores do not answer whether a proposed release is supported for a named use, population, and threat model. We introduce SynthGuard-ReleaseBench, an audit framework that locks the use, candidate panel, tolerances, and audit schedule before evaluation. It compares real-trained and synthetic-trained workflows on protected data, gives simultaneous finite-sample bounds for bounded loss gaps, requires controls, and keeps utility, empirical privacy risk, mechanism claims, and human release authority separate. Across four American Community Survey studies, five non-ACS records, two chronological diagnostics, and a sealed prototype, the benchmark retains favorable, unfavorable, and excluded outcomes. Transparent baselines pass some locked audits; compact learned models fail under the declared budgets; a health-table case is excluded because its negative control passes. A post-audit scaling arm, repeated across three generation seeds, shows the same locked criterion admitting those learned models once they are fit on enough data while still rejecting a dependence-destroying control at every size, so the criterion discriminates rather than merely rejects; the same repetition withdraws a finer single-seed ordering. The theory adds a pre-audit sample-size rule, variance-adaptive and anytime-valid certificates that tighten the bound two to ten times on the same locked evidence, a temporal certificate for time-ordered audits, and two lower bounds: ordinary bounded queries reconstruct a protected audit once the query budget reaches its size, and the panel-size correction is necessary rather than conservative. The contribution is a reproducible workflow for use-specific release evidence, not a claim that any generator is private, safe, or deployment-ready.
Related
- Beyond Distributional Fidelity: Causal-Penalized Diffusion for Synthetic Tabular Data
- Seq2Synth: Benchmarking Temporal Fidelity in Synthetic Sequential Tabular Data
- Challenges of analyzing synthetic tabular data generated from 115 phase 3 oncology trials
- Classical methods for modern applications: synthetic tabular data generation using lightweight discrete response continuous latent variable model
- Estimating Uncoded Crash Factors with Tabular Foundation and System One Models: Kumo Tabular and Jev
BibTeX
@misc{opoku2026synthguardreleasebench,
title = {{SynthGuard-ReleaseBench: Locked-Audit Evidence for Synthetic Tabular Data Releases}},
author = {Jeffery Opoku and David Banahene},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2608.14753},
eprint = {2608.14753},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.14753},
}
RIS
TY - UNPB
TI - SynthGuard-ReleaseBench: Locked-Audit Evidence for Synthetic Tabular Data Releases
AU - Jeffery Opoku
AU - David Banahene
PY - 2026
DA - 2026/10/27
PB - arXiv
DO - 10.48550/arxiv.2608.14753
UR - https://arxiv.org/abs/2608.14753
AB - Synthetic tabular data are often judged by realism, privacy, or downstream-task scores. Those scores do not answer whether a proposed release is supported for a named use, population, and threat model. We introduce SynthGuard-ReleaseBench, an audit framework that locks the use, candidate panel, tolerances, and audit schedule before evaluation. It compares real-trained and synthetic-trained workflows on protected data, gives simultaneous finite-sample bounds for bounded loss gaps, requires controls, and keeps utility, empirical privacy risk, mechanism claims, and human release authority separate. Across four American Community Survey studies, five non-ACS records, two chronological diagnostics, and a sealed prototype, the benchmark retains favorable, unfavorable, and excluded outcomes. Transparent baselines pass some locked audits; compact learned models fail under the declared budgets; a health-table case is excluded because its negative control passes. A post-audit scaling arm, repeated across three generation seeds, shows the same locked criterion admitting those learned models once they are fit on enough data while still rejecting a dependence-destroying control at every size, so the criterion discriminates rather than merely rejects; the same repetition withdraws a finer single-seed ordering. The theory adds a pre-audit sample-size rule, variance-adaptive and anytime-valid certificates that tighten the bound two to ten times on the same locked evidence, a temporal certificate for time-ordered audits, and two lower bounds: ordinary bounded queries reconstruct a protected audit once the query budget reaches its size, and the panel-size correction is necessary rather than conservative. The contribution is a reproducible workflow for use-specific release evidence, not a claim that any generator is private, safe, or deployment-ready.
ER -