FairCauseSyn: Towards Causally Fair LLM-Augmented Synthetic Data Generation

• URL: http://arxiv.org/abs/2506.19082v1
• Date: Mon, 23 Jun 2025 19:59:26 GMT
• Title: FairCauseSyn: Towards Causally Fair LLM-Augmented Synthetic Data Generation
• Authors: Nitish Nagesh, Ziyu Wang, Amir M. Rahmani,
• Abstract summary: Synthetic data generation creates data based on real-world data using generative models.<n>We develop the first LLM-augmented synthetic data generation method to enhance causal fairness using real-world health data.<n>When trained on causally fair predictors, synthetic data reduces bias on the sensitive attribute by 70% compared to real data.
• Score: 4.392938909804638
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Synthetic data generation creates data based on real-world data using generative models. In health applications, generating high-quality data while maintaining fairness for sensitive attributes is essential for equitable outcomes. Existing GAN-based and LLM-based methods focus on counterfactual fairness and are primarily applied in finance and legal domains. Causal fairness provides a more comprehensive evaluation framework by preserving causal structure, but current synthetic data generation methods do not address it in health settings. To fill this gap, we develop the first LLM-augmented synthetic data generation method to enhance causal fairness using real-world tabular health data. Our generated data deviates by less than 10% from real data on causal fairness metrics. When trained on causally fair predictors, synthetic data reduces bias on the sensitive attribute by 70% compared to real data. This work improves access to fair synthetic data, supporting equitable health research and healthcare delivery.

Related papers

• Investigation of Accuracy and Bias in Face Recognition Trained with Synthetic Data [10.241047069730058]
  We evaluate the impact of synthetic data on bias and performance of face recognition systems.<n>By maintaining equal identity count across synthetic and real datasets, we ensure fair comparisons.<n>Our results demonstrate that although synthetic data still lags behind the real datasets in the generalization on IJB-B/C, demographically balanced synthetic datasets, especially those generated with SD35, show potential for bias mitigation.
  arXiv  Detail & Related papers  (2025-07-28T12:52:23Z)
• AIM-Fair: Advancing Algorithmic Fairness via Selectively Fine-Tuning Biased Models with Contextual Synthetic Data [44.94133254226272]
  Existing methods often face limitations in the diversity and quality of synthetic data, leading to compromised fairness and overall model accuracy.<n>This paper proposes AIM-Fair, aiming to overcome these limitations and harness the potential of cutting-edge generative models in promoting algorithmic fairness.<n> Experiments on CelebA and UTKFace datasets show that our AIM-Fair improves model fairness while maintaining utility, outperforming both fully and partially fine-tuned approaches to model fairness.
  arXiv  Detail & Related papers  (2025-03-07T18:26:48Z)
• Can Synthetic Data be Fair and Private? A Comparative Study of Synthetic Data Generation and Fairness Algorithms [2.144088660722956]
  We find that the DEbiasing CAusal Fairness (DECAF) algorithm achieves the best balance between privacy and fairness.<n>Applying pre-processing fairness algorithms to synthetic data improves fairness even more than when applied to real data.
  arXiv  Detail & Related papers  (2025-01-03T12:35:58Z)
• Little Giants: Synthesizing High-Quality Embedding Data at Scale [71.352883755806]
  We introduce SPEED, a framework that aligns open-source small models to efficiently generate large-scale embedding data. SPEED uses only less than 1/10 of the GPT API calls, outperforming the state-of-the-art embedding model E5_mistral when both are trained solely on their synthetic data.
  arXiv  Detail & Related papers  (2024-10-24T10:47:30Z)
• Improving Grammatical Error Correction via Contextual Data Augmentation [49.746484518527716]
  We propose a synthetic data construction method based on contextual augmentation. Specifically, we combine rule-based substitution with model-based generation. We also propose a relabeling-based data cleaning method to mitigate the effects of noisy labels in synthetic data.
  arXiv  Detail & Related papers  (2024-06-25T10:49:56Z)
• Best Practices and Lessons Learned on Synthetic Data [83.63271573197026]
  The success of AI models relies on the availability of large, diverse, and high-quality datasets. Synthetic data has emerged as a promising solution by generating artificial data that mimics real-world patterns.
  arXiv  Detail & Related papers  (2024-04-11T06:34:17Z)
• Strong statistical parity through fair synthetic data [0.0]
  This paper explores the creation of synthetic data that embodies Fairness by Design. A downstream model trained on such synthetic data provides fair predictions across all thresholds.
  arXiv  Detail & Related papers  (2023-11-06T10:06:30Z)
• Boosting Data Analytics With Synthetic Volume Expansion [3.568650932986342]
  This article explores the effectiveness of statistical methods on synthetic data and the privacy risks of synthetic data. A key finding within this framework is the generational effect, which reveals that the error rate of statistical methods on synthetic data decreases with the addition of more synthetic data but may eventually rise or stabilize.
  arXiv  Detail & Related papers  (2023-10-27T01:57:27Z)
• Synthetic data, real errors: how (not) to publish and use synthetic data [86.65594304109567]
  We show how the generative process affects the downstream ML task. We introduce Deep Generative Ensemble (DGE) to approximate the posterior distribution over the generative process model parameters.
  arXiv  Detail & Related papers  (2023-05-16T07:30:29Z)
• Downstream Fairness Caveats with Synthetic Healthcare Data [21.54509987309669]
  Privacy laws limit access to health data such as Electronic Medical Records (EMRs) to preserve patient privacy. This paper evaluates synthetically generated healthcare data for biases and investigates the effect of fairness mitigation techniques on utility-fairness.
  arXiv  Detail & Related papers  (2022-03-09T00:52:47Z)
• DECAF: Generating Fair Synthetic Data Using Causally-Aware Generative Networks [71.6879432974126]
  We introduce DECAF: a GAN-based fair synthetic data generator for tabular data. We show that DECAF successfully removes undesired bias and is capable of generating high-quality synthetic data. We provide theoretical guarantees on the generator's convergence and the fairness of downstream models.
  arXiv  Detail & Related papers  (2021-10-25T12:39:56Z)
• Representative & Fair Synthetic Data [68.8204255655161]
  We present a framework to incorporate fairness constraints into the self-supervised learning process. We generate a representative as well as fair version of the UCI Adult census data set. We consider representative & fair synthetic data a promising future building block to teach algorithms not on historic worlds, but rather on the worlds that we strive to live in.
  arXiv  Detail & Related papers  (2021-04-07T09:19:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.