Tabular Transformers Meet Relational Databases
ACM Transactions on Intelligent Systems and Technology · 31.10.2025 · 16 (5) · 1-24
DOI: 10.1145/3749991 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Intelligent Systems and Technology, online erschienen am 18.09.2025. Verfasst von Jakub Peleška und 1 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 72 Literaturangaben. Gefördert von: Czech Science Foundation, European Union’s Sponsor Horizon Europe Research and Innovation Program.
Abstract (Originalsprache)
Transformer models have continuously expanded into all machine learning domains convertible to the underlying sequence-to-sequence representation, including tabular data. However, while ubiquitous, this representation restricts its extension to the more general case of relational databases . In this article, we introduce a modular neural message-passing scheme that closely adheres to the formal relational model, enabling direct end-to-end learning of tabular transformers from database storage systems. We address the associated challenges of appropriate learning data representation and loading, which are critical in the database setting, and compare our approach against a number of representative models from various related fields across a significantly wide range of datasets. Our results then demonstrate superior performance of this newly proposed class of neural architectures.
Förderung
- Czech Science Foundation
- European Union’s Sponsor Horizon Europe Research and Innovation Program
Zitieren
APA
Peleška, J., & Šír, G. (2025). Tabular Transformers Meet Relational Databases. ACM Transactions on Intelligent Systems and Technology, 16(5), 1-24. https://doi.org/10.1145/3749991
Vancouver
Peleška J, Šír G. Tabular Transformers Meet Relational Databases. ACM Transactions on Intelligent Systems and Technology. 2025;16(5):1-24. doi:10.1145/3749991
BibTeX
@article{Peleka2025,
author = {Peleška, Jakub and Šír, Gustav},
title = {Tabular Transformers Meet Relational Databases},
journal = {ACM Transactions on Intelligent Systems and Technology},
year = {2025},
volume = {16},
number = {5},
pages = {1--24},
doi = {10.1145/3749991}
}
EUVIMED