7 papers
CDMD: A Cross-Dataset Mixed-Type Diffusion Model for Tabular Data
Mohamed Amine Ketata, Maximilian Schambach, Stephan Günnemann
Generative models for tabular data are typically trained separately for each dataset, limiting knowledge transfer and requiring the storage of many specialized models. In this pape…
Benchmarking Attention for Tabular Foundation Models
Maximilian Schambach, Clemens Biehl, Sam Thelin
Tabular in-context learners such as TabPFN, Mitra, or ConTextTab rely on alternating row and column attention over 2D sequences of latent embeddings. These attention patterns diffe…
Enhancing Tabular Learners with Context-Aware Semantic Embeddings
Günther Schindler, Maximilian Schambach, Johannes Höhne
While modern tabular learners excel at capturing statistical patterns, they frequently operate in a semantic vacuum, treating textual features as discrete symbols, ignoring the ric…
FlexTab: A Flexible Encoder-Decoder Architecture for In-Context Learning Across Diverse Tabular Tasks
Marek Polewczyk, Maximilian Schambach, Marco Spinaci +2
We introduce FlexTab, a flexible encoder-decoder architecture for in-context learning on tabular data that pairs a single, task-agnostic encoder with a suite of task-specific decod…
Exploring Differences Between Tabular Enterprise Data and Public Benchmarks
Myung Jun Kim, Maximilian Schambach, Frank Essenberger +2
Tabular data dominate the landscape of data science, increasingly attracting innovative machine learning models and tailored benchmarks. Yet, little is known for enterprise data, w…
TabGemma: Text-Based Tabular ICL via LLM using Continued Pretraining and Retrieval
Günther Schindler, Maximilian Schambach, Michael Medek +1
We study LLMs for tabular prediction with mixed text, numeric, and categorical fields. We introduce TabGemma, a schema-agnostic in-context learner that treats rows as sequences and…
ConTextTab: A Semantics-Aware Tabular In-Context Learner
Marco Spinaci, Marek Polewczyk, Maximilian Schambach +1
Tabular in-context learning (ICL) has recently achieved state-of-the-art (SOTA) performance on several tabular prediction tasks. Previously restricted to classification problems on…