# Beyond Distributional Fidelity: Causal-Penalized Diffusion for Synthetic Tabular Data

arXiv: https://arxiv.org/abs/2610.11407 · HTML: https://mlsanity.com/paper/2610.11407/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 26 references that are in the map.
- First posted on arXiv 8 Oct 2026.

## Builds on (most cited first, 10 of 10)

- [Language Models are Realistic Tabular Data Generators](https://mlsanity.com/paper/2210.06280/) (2022)
- [Data Synthesis based on Generative Adversarial Networks](https://mlsanity.com/paper/1806.03384/) (2018)
- [Synthetic Data -- what, why and how?](https://mlsanity.com/paper/2205.03257/) (2022)
- [CausalPFN: Amortized Causal Effect Estimation via In-Context Learning](https://mlsanity.com/paper/2506.07918/) (2025)
- [A Survey on Deep Learning Approaches for Tabular Data Generation: Utility, Alignment, Fidelity, Privacy, Diversity, and Beyond](https://mlsanity.com/paper/2503.05954/) (2025)
- [Robust Variational Autoencoder for Tabular Data with Beta Divergence](https://mlsanity.com/paper/2006.08204/) (2020)
- [Evaluating Generative Models for Tabular Data: Novel Metrics and Benchmarking](https://mlsanity.com/paper/2504.20900/) (2025)
- [CausalWrap: Model-Agnostic Causal Constraint Wrappers for Tabular Synthetic Data](https://mlsanity.com/paper/2603.02015/) (2026)
- [Generative Synthetic Data for Causal Inference: Pitfalls, Remedies, and Opportunities](https://mlsanity.com/paper/2604.23904/) (2026)
- [ReTabSyn: Realistic Tabular Data Synthesis via Reinforcement Learning](https://mlsanity.com/paper/2603.10823/) (2026)

## Abstract (arXiv)

Synthetic tabular generators are commonly optimized for distributional fidelity, but statistical similarity alone does not guarantee preservation of causal effects. In this paper, we study whether causal fidelity can be improved directly within a fully generative tabular model. Causal Fidelity is defined with respect to a target estimand as the discrepancy between inferential distributions obtained from real and synthetic data, and theoretical results show that high statistical fidelity does not generally imply high causal fidelity. We then propose a causal-fidelity-aware training framework which adds a causal discrepancy penalty to the generative objective. The framework is instantiated with a causal-penalized TabDDPM and optimized using an on-policy score-function estimator. We further establish conditions under which causal regularization improves expected causal fidelity. Experiments across diverse treatment-effect simulations and two benchmark datasets evaluate the ability of our method to improve causal fidelity while preserving competitive statistical fidelity.