Memory Efficient Tabular Foundation Models

Shuting Luo, Monika Mikhail Kanaan, Cameron Gordon, Anna Leontjeva, Simon Lucey

Keywords: Tabular Foundation Models, Quantization, Memory-Efficiency

TL;DR: We reduce the memory footprint of Tabular Foundation Models for practical weight deployment.

Abstract: Tabular Foundation Models, such as TabPFN, have received a large amount of recent attention due to their performance on in-context tabular machine learning tasks, which often exceeds classical baselines. However, practical deployment considerations of these models has received less attention. In this paper we investigate the memory requirements for these models. We demonstrate that employing model compression approaches can enable memory reductions of up to 7.6$\times$ with similar levels of performance, reducing deployment requirements by nearly 87%. Our work provides insight to practitioners seeking efficient deployment of these models in practical settings.

Submission Number: 197