Data as of Sep 29, 2026 · Based on 166 AI responses · See how Parse measures this
TabSyn is a deep generative model that synthesizes mixed-type tabular data (continuous and categorical) using a variational autoencoder and a score-based diffusion model in latent space. It achieves state-of-the-art performance in recovering ground truth distributions and offers faster sampling than previous diffusion-based methods.
Hosted on GitHub
0%No change
of AI answers about TabSyn and its rivals. Week of Sep 21
AI mentioned TabSyn in 0% of answers about TabSyn and its rivals in the week of Sep 21.
