CausalMixFT Versus Diffusion-Based Augmentation for Robust Tabular Foundation Model Fine-Tuning Under Covariate Shift

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: Does CausalMixFT outperform diffusion-based data augmentation (e.g., DiffAugment) in terms of robustness to covariate shift when fine-tuning tabular foundation models, evaluated using accuracy on out-of-distribution splits of TabMWP?

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