MaskTab: Scalable Masked Tabular Pretraining with Scaling Laws and Distillation for Industrial Classification

TL;DR

MaskTab introduces a scalable, self-supervised pretraining framework for industrial tabular data, effectively handling missing values and enabling robust, lightweight models for high-stakes decision systems.

Contribution

It presents a novel unified pretraining approach tailored for complex tabular data, incorporating missing value encoding, hybrid training schemes, and model distillation.

Findings

Achieves +5.04% AUC and +8.28% KS improvements on benchmarks.

Distills into lightweight models with +2.55% AUC and +4.85% KS gains.

Enhances robustness to distribution shifts in industrial datasets.

Abstract

Tabular data forms the backbone of high-stakes decision systems in finance, healthcare, and beyond. Yet industrial tabular datasets are inherently difficult: high-dimensional, riddled with missing entries, and rarely labeled at scale. While foundation models have revolutionized vision and language, tabular learning still leans on handcrafted features and lacks a general self-supervised framework. We present MaskTab, a unified pre-training framework designed specifically for industrial-scale tabular data. MaskTab encodes missing values via dedicated learnable tokens, enabling the model to distinguish structural absence from random dropout. It jointly optimizes a hybrid supervised pre-training scheme--utilizing a twin-path architecture to reconcile masked reconstruction with task-specific supervision--and an MoE-augmented loss that adaptively routes features through specialized subnetworks. On industrial-scale benchmarks, it achieves +5.04% AUC and +8.28% KS over prior art under rigorous scaling. Moreover, its representations distill effectively into lightweight models, yielding +2.55% AUC and +4.85% KS under strict latency and interpretability constraints, while improving robustness to distribution shifts. Our work demonstrates that tabular data admits a foundation-model treatment--when its structural idiosyncrasies are respected.