Correlation of Structural Fidelity Metrics with Downstream Performance in Tabular Generative Model Pretraining

Generative models have revolutionized multiple domains, yet their application to tabular data remains underexplored. Evaluating generative models for tabular data presents unique challenges due to structural complexity, large-scale variability, and mixed data types, making it difficult to intuitively capture intricate patterns. Existing evaluation metrics offer only partial insights, lacking a comprehensive measure of generative performance. To address this limitation, we propose three novel evaluation metrics: FAED, FPCAD, and RFIS. Our extensive experimental analysis, conducted on three stan

Research goal: How do novel structural fidelity metrics for tabular generative models correlate with downstream performance when used to pre-train large language models on mixed-type data?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.