CausalMixFT Robustness Under Varying Synthetic Pretraining Dataset Sizes on TabFact

Fine-tuning tabular foundation models (TFMs) under data scarcity is challenging, as early stopping on even scarcer validation data often fails to capture true generalization performance. We propose CausalMixFT, a method that enhances fine-tuning robustness and downstream performance by generating structurally consistent synthetic samples using Structural Causal Models (SCMs) fitted on the target dataset. This approach augments limited real data with causally informed synthetic examples, preserving feature dependencies while expanding training diversity. Evaluated across 33 classification datas

Research goal: What is the impact of varying the size of synthetic pretraining datasets on the robustness of tabular foundation models trained with CausalMixFT, measured by accuracy on the TabFact benchmark and compared to models trained with conventional data augmentation methods?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 7.6/10.