Comparative Robustness of CausalMixFT Synthetic Data for Fine-Tuning Tabular Foundation Models on TabMWP and TabStruct

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 comparative robustness of CausalMixFT-generated synthetic data against other data augmentation methods (e.g., GAN-based or diffusion-based) for fine-tuning tabular foundation models, measured by validation accuracy and downstream task performance on TabMWP and TabStruct?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.7/10.