Breaking the Quality-Privacy Tradeoff in Tabular Data Generation via In-Context Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917464182358016 |
|---|---|
| author | Han, Xinyan Lu, Yan Lin, Xiaoyu Jiang, Yuanyuan Wang, Yuanrui Li, Xuanyue Zou, Wenchao Zhang, Xingxuan |
| author_facet | Han, Xinyan Lu, Yan Lin, Xiaoyu Jiang, Yuanyuan Wang, Yuanrui Li, Xuanyue Zou, Wenchao Zhang, Xingxuan |
| contents | Tabular data synthesis aims to generate high-quality data while preserving privacy. However, we find that existing tabular generative models exhibit a clear tradeoff in the small-data regime: improving data quality typically comes at the cost of increased memorization of training samples, thereby weakening privacy protection. This tradeoff arises because small training sets make it difficult for dataset-specific generative models to distinguish generalizable structure from sample-specific patterns. To address this, we propose DiffICL, which formulates tabular data generation as an in-context learning problem. Instead of fitting each dataset from scratch,DiffICL leverages pretrained structural priors learned from a large collection of datasets, enabling it to infer data distributions from limited context rather than memorizing individual samples. We evaluate DiffICL on 14 real-world datasets. Results show that DiffICL improves both data quality and privacy, and generate synthetic data that provides effective data augmentation. Our findings suggest that the quality-privacy tradeoff can be improved through better training paradigms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_04911 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Breaking the Quality-Privacy Tradeoff in Tabular Data Generation via In-Context Learning Han, Xinyan Lu, Yan Lin, Xiaoyu Jiang, Yuanyuan Wang, Yuanrui Li, Xuanyue Zou, Wenchao Zhang, Xingxuan Machine Learning Tabular data synthesis aims to generate high-quality data while preserving privacy. However, we find that existing tabular generative models exhibit a clear tradeoff in the small-data regime: improving data quality typically comes at the cost of increased memorization of training samples, thereby weakening privacy protection. This tradeoff arises because small training sets make it difficult for dataset-specific generative models to distinguish generalizable structure from sample-specific patterns. To address this, we propose DiffICL, which formulates tabular data generation as an in-context learning problem. Instead of fitting each dataset from scratch,DiffICL leverages pretrained structural priors learned from a large collection of datasets, enabling it to infer data distributions from limited context rather than memorizing individual samples. We evaluate DiffICL on 14 real-world datasets. Results show that DiffICL improves both data quality and privacy, and generate synthetic data that provides effective data augmentation. Our findings suggest that the quality-privacy tradeoff can be improved through better training paradigms. |
| title | Breaking the Quality-Privacy Tradeoff in Tabular Data Generation via In-Context Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.04911 |