TabDDPM
PulseAugur coverage of TabDDPM — every cluster mentioning TabDDPM across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Deep tabular generative models fail to outperform baselines in benchmark study
A new benchmark study on arXiv investigates the effectiveness of deep tabular generative models compared to simpler baselines. Across 8 public datasets subsampled from 200 to 20,000 rows, and 4 natively small clinical d…
-
New framework PEG-Tab reduces data synthesis risks for tabular generators
Researchers have developed PEG-Tab, a post-training framework designed to repair and control the release of synthesized tabular data. This method aims to mitigate risks associated with pretrained tabular generators that…
-
New framework uses diffusion models to synthesize realistic cyber-physical attacks for 5G systems
Researchers have developed Diff-DDoS, a novel framework designed to improve the detection of cyber-physical attacks in 5G-enabled systems. This framework utilizes tabular diffusion models to synthesize realistic attack …
-
New FairDiffuseVQVAE model enhances synthetic data fairness without sacrificing quality
Researchers have developed FairDiffuseVQVAE, a novel two-stage architecture for generating synthetic tabular data that decouples data fidelity from fairness. The first stage uses a vector-quantized autoencoder for recon…