TabTune: A Unified Library for Inference and Fine-Tuning Tabular Foundation Models
This addresses the adoption barriers for tabular foundation models in structured data learning, though it is incremental as it builds on existing models and methods.
The authors tackled the problem of fragmented and inconsistent workflows for tabular foundation models by developing TabTune, a unified library that standardizes preprocessing, fine-tuning, and evaluation, resulting in support for seven state-of-the-art models and multiple adaptation strategies.
Tabular foundation models represent a growing paradigm in structured data learning, extending the benefits of large-scale pretraining to tabular domains. However, their adoption remains limited due to heterogeneous preprocessing pipelines, fragmented APIs, inconsistent fine-tuning procedures, and the absence of standardized evaluation for deployment-oriented metrics such as calibration and fairness. We present TabTune, a unified library that standardizes the complete workflow for tabular foundation models through a single interface. TabTune provides consistent access to seven state-of-the-art models supporting multiple adaptation strategies, including zero-shot inference, meta-learning, supervised fine-tuning (SFT), and parameter-efficient fine-tuning (PEFT). The framework automates model-aware preprocessing, manages architectural heterogeneity internally, and integrates evaluation modules for performance, calibration, and fairness. Designed for extensibility and reproducibility, TabTune enables consistent benchmarking of adaptation strategies of tabular foundation models.