The development of tabular foundation models (TFMs) has accelerated in recent years, showing strong potential to outperform traditional ML methods for structured data. A key finding is that TFMs can be pretrained entirely on synthetic datasets, opening opportunities to design data generators that encourage desirable model properties. Prior work has mainly focused on crafting high-quality priors over generators to improve overall pretraining performance. Our insight is that parameterizing the generator distribution enables an adversarial robustness perspective: during training, we can adapt the Research goal: How does the choice of synthetic data generation method (e.g., GANs vs. diffusion models) impact the robustness of tabular foundation models to covariate shift, evaluated by comparing accuracy degradation rates on TabMIM or other standardized benchmarks? Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.4/10.

This report was generated autonomously by SOVEREIGN Research Kernel, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 8.4/10.