Constrained Tabular Diffusion for Finance
Name
3768292.3770358.pdf
Size
1.64 MB
Format
Adobe PDF
Checksum (MD5)
3723fd6f7dbddc2526ab7b9d056ef6be
Author(s)
Cardei, Michael
Munoz, Jose
Barrera, Oscar
Chandrahas, Shreyas
Saha, Partha
Date Issued
November 14, 2025
Publisher
ACM|6th ACM International Conference on AI in Finance
Citation
Michael Cardei, Jose Munoz, Oscar Barrera, Shreyas Chandrahas, and Partha Saha. 2025. Constrained Tabular Diffusion for Finance. In Proceedings of the 6th ACM International Conference on AI in Finance (ICAIF '25). Association for Computing Machinery, New York, NY, USA, 543–551.
Version
Final published version
Abstract
Generative models in finance face the dual challenge of producing realistic data while satisfying strict regulatory and economic objectives, a requirement that standard tabular diffusion models cannot provide. To address this difficulty, we introduce Constrained Tabular Diffusion for Finance (CTDF), a novel integration of sampling-time feasibility operations with mixed-type tabular diffusion in financial applications. By incorporating a training-free feasibility operator into the reverse‑diffusion sampling loop, CTDF enforces hard constraints for applications such as simulation, legal compliance, and extrapolation. Extensive experiments on large-scale financial datasets demonstrate zero constraint violations and improvement in scarce data utility. CTDF establishes a robust method for generating trustworthy and compliant synthetic data, opening new avenues for rigorous generative modeling and analysis in the financial domain.
Description
6th ACM International Conference on AI in Finance (ICAIF ’25), November 15–18, 2025,
Singapore, Singapore
MIT Department
Massachusetts Institute of Technology. Department of Physics
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3768292.3770358
