activity
20242026
collaborators
Showing cs.LGShow all

5 papers · 1 filter

cs.LG2026

Statistically Indistinguishable, Operationally Distinct: A Formal Barrier for Tabular Foundation Models

Tassilo Klein, Johannes Hoffart

Tabular foundation models cannot reason about data produced by running systems without access to the rules that govern them. We make this statement falsifiable. The \emph{Operation…

cs.LG2026

Tabular Foundation Model for Generative Modelling

Xiangjian Jiang, Mingxuan Liu, Nikola Simidjievski +2

Generative modelling is a demanding test of foundation models, because it requires robust, holistic representation learning for a given data modality, rather than optimisation for…

cs.LG2026

Position: Foundation Models for Tabular Data within Systemic Contexts Need Grounding

Tassilo Klein, Johannes Hoffart

This position paper argues that foundation models for tabular data face inherent limitations when isolated from operational context - the procedural logic, declarative rules, and d…

cs.LG2025

SALT: Sales Autocompletion Linked Business Tables Dataset

Tassilo Klein, Clemens Biehl, Margarida Costa +3

Foundation models, particularly those that incorporate Transformer architectures, have demonstrated exceptional performance in domains such as natural language processing and image…

cs.LG2024

PORTAL: Scalable Tabular Foundation Models via Content-Specific Tokenization

Marco Spinaci, Marek Polewczyk, Johannes Hoffart +3

Self-supervised learning on tabular data seeks to apply advances from natural language and image domains to the diverse domain of tables. However, current techniques often struggle…