DEREC-SIMPRO: unlock Language Model benefits to advance Synthesis in Data Clean Room
Not on the map. Why?
Abstract (arXiv)
Data collaboration via Data Clean Room offers value but raises privacy concerns, which can be addressed through synthetic data and multi-table synthesizers. Common multi-table synthesizers fail to perform when subjects occur repeatedly in both tables. This is an urgent yet unresolved problem, since having both tables with repeating subjects is common. To improve performance in this scenario, we present the DEREC 3-step pre-processing pipeline to generalize adaptability of multi-table synthesizers. We also introduce the SIMPRO 3-aspect evaluation metrics, which leverage conditional distribution and large-scale simultaneous hypothesis testing to provide comprehensive feedback on synthetic data fidelity at both column and table levels. Results show that using DEREC improves fidelity, and multi-table synthesizers outperform single-table counterparts in collaboration settings. Together, the DEREC-SIMPRO pipeline offers a robust solution for generalizing data collaboration, promoting a more efficient, data-driven society.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 28 of its 59 references that are in the map. First posted on arXiv 31 Oct 2024.
Builds on
28 of 59 references.
References · 28
- 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
- 2023Extracting Training Data from Diffusion Modelscited by 723
- 2019Modeling Tabular data using Conditional GANcited by 695
- 2019Confident Learning: Estimating Uncertainty in Dataset Labelscited by 411
- 2022TabDDPM: Modelling Tabular Data with Diffusion Modelscited by 373
- 2022Language Models are Realistic Tabular Data Generatorscited by 290
- 2018Statistical Aspects of Wasserstein Distancescited by 202
- 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 199
- 2017Generating Multi-label Discrete Patient Records using Generative Adversarial Networkscited by 197
- 2018Data Synthesis based on Generative Adversarial Networkscited by 184
- 2015On Wasserstein Two Sample Testing and Related Families of Nonparametric Testscited by 165
- 2021CTAB-GAN: Effective Table Data Synthesizingcited by 156
16 more
- 2023REaLTabFormer: Generating Realistic Relational and Tabular Data using Transformerscited by 93
- 2020Tabular Transformers for Modeling Multivariate Time Seriescited by 77
- 2020A Non-Parametric Test to Detect Data-Copying in Generative Modelscited by 52
- 2020Online Forgetting Process for Linear Regression Modelscited by 21
- 2021Fidelity and Privacy of Synthetic Medical Datacited by 16
- 2024Data Plagiarism Index: Characterizing the Privacy Risk of Data-Copying in Tabular Generative Modelscited by 12
- 2021Fund2Vec: Mutual Funds Similarity using Graph Learningcited by 11
- 2024SiloFuse: Cross-silo Synthetic Data Generation with Latent Tabular Diffusion Modelscited by 10
- 2024Discriminative Estimation of Total Variation Distance: A Fidelity Auditor for Generative Datacited by 10
- 2022On the Utility Recovery Incapability of Neural Net-based Differential Private Tabular Training Data Synthesizer under Privacy Deregulationcited by 8
- 2024Downstream Task-Oriented Generative Model Selections on Synthetic Data Training for Fraud Detection Modelscited by 7
- 2022Modular Regression: Improving Linear Models by Incorporating Auxiliary Datacited by 5
- 2024Improve Fidelity and Utility of Synthetic Credit Card Transaction Time Series from Data-centric Perspectivecited by 5
- 2024BadGD: A unified data-centric framework to identify gradient descent vulnerabilitiescited by 5
- 2024Advancing Retail Data Science: Comprehensive Evaluation of Synthetic Datacited by 5
- 2024Data Deletion for Linear Regression with Noisy SGDcited by 2
Built on by
Citing papers per year: 2025: 2.
2 papers.