aiqbench analysis
CDMD advances tabular data generation by unifying numerical and categorical features into a single diffusion framework, enabling efficient knowledge transfer across heterogeneous datasets without requiring separate models for each schema. This approach significantly reduces parameter overhead while improving generation quality on unseen data through pre-training on large corpora of diverse datasets. The work addresses a critical bottleneck in applying generative AI to structured data by solving the challenge of mixed-type representation and cross-dataset adaptability.
CDMD: A Cross-Dataset Mixed-Type Diffusion Model for Tabular Data
arXiv:2609.39124v1 Announce Type: new Abstract: Generative models for tabular data are typically trained separately for each dataset, limiting knowledge transfer and requiring the storage of many specialized models. In this paper, we introduce CDMD, a tabular diffusion model trained jointly across heterogeneous datasets with different schemas and variable numbers of numerical and categorical features. Unlike existing cross-dataset tabular diffusion models that operate in continuous representation spaces, CDMD defines diffusion directly over the mixed-type feature space and is trained end-to-end. To accommodate heterogeneous categorical domains, we introduce a schema-restricted reverse-process parameterization for masked diffusion models, in which the output space dynamically adapts to each feature's vocabulary. We then compose numerical and categorical feature-level diffusion processes into a schema-dependent row-level process. A shared schema-aware Transformer denoiser captures dependencies between features and parameterizes the reverse process across varying schemas. On seven real-world datasets, a single jointly trained CDMD achieves the highest average generation quality among strong single-dataset and cross-dataset baselines, while using substantially fewer total parameters than the collection of separately trained models. Furthermore, pre-training on a corpus of 337 datasets improves generation on previously unseen datasets under both limited target data and limited adaptation epochs. These results demonstrate the potential of direct mixed-type diffusion for shared and transferable tabular data generation. Our code is available at https://github.com/ketatam/cdmd.
Abstract:Generative models for tabular data are typically trained separately for each dataset, limiting knowledge transfer and requiring the storage of many specialized models. In this paper, we introduce CDMD, a tabular diffusion model trained jointly across heterogeneous datasets with different schemas and variable numbers of numerical and categorical features. Unlike existing cross-dataset tabular diffusion models that operate in continuous representation spaces, CDMD defines diffusion directly over the mixed-type feature space and is trained end-to-end. To accommodate heterogeneous categorical domains, we introduce a schema-restricted reverse-process parameterization for masked diffusion models, in which the output space dynamically adapts to each feature's vocabulary. We then compose numerical and categorical feature-level diffusion processes into a schema-dependent row-level process. A shared schema-aware Transformer denoiser captures dependencies between features and parameterizes the reverse process across varying schemas. On seven real-world datasets, a single jointly trained CDMD achieves the highest average generation quality among strong single-dataset and cross-dataset baselines, while using substantially fewer total parameters than the collection of separately trained models. Furthermore, pre-training on a corpus of 337 datasets improves generation on previously unseen datasets under both limited target data and limited adaptation epochs. These results demonstrate the potential of direct mixed-type diffusion for shared and transferable tabular data generation. Our code is available at this https URL.
Machine Learning (cs.LG)
arXiv:2609.39124 [cs.LG]
(or arXiv:2609.39124v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.39124 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 306 words
Related coverage
SurvDiff: A Diffusion Model for Generating Synthetic Data in Survival Analysis
arXiv · Jul 28, 2026
Logit-Coordinate Generative Models for Mixed Continuous-Categorical Tabular Data
arXiv · Jul 28, 2026
MIND: Marginal-Invariant Neural Dependency Diffusion for Mixed-Type Tabular Generation
arXiv · Oct 1, 2026
Multimodal synthesis of MRI and tabular data with diffusion in a joint latent space via cross-attention
arXiv · Oct 15, 2026