Paper on Foundation Models for Relational Databases accepted to TRL@NeurIPS 2022

Congratulations to Liane Vogel, Benjamin Hilprecht and Carsten Binnig!


Liane Vogel will present the paper “Towards Foundation Models for Relational Databases” at Table Representation Learning Workshop @ NeurIPS 2022 on December 2nd in New Orleans.

Tabular representation learning has recently gained a lot of attention. However, existing approaches only learn a representation from a single table, and thus ignore the potential to learn from the full structure of relational databases, including neighboring tables that can contain important information for a contextualized representation. Moreover, current models are significantly limited in scale, which prevents that they learn from large databases.

In this paper, we thus introduce ourvision of relational representation learning, that can not only learn from the full relational structure, but also can scale to larger database sizes that are commonly found in real-world. Moreover, we also discuss opportunities and challenges we see along the way to enable this vision and present initial very promising results.

Overall, we argue that this direction can lead to foundation models for relational databases that are today only available for text and images.