cs.LGNov 9, 2026

Out-of-the-Box Tabular Foundation Models for Bayesian Optimization in Materials Discovery

Authors: Klemens Flöge, Simon Bing, Leo Grinsztajn, Eliott Kalfon, Lennart Purucker, Noah Hollmann, Frank Hutter

Organizations: Prior Labs · University of Freiburg · ELLIS Institute Tübingen

Abstract

Bayesian optimization is widely used in active learning loops for materials science. Gaussian processes serve as the de-facto surrogate model family, but require task-dependent kernel choices and can lack the modeling capacity for complex relationships. Tabular foundation models (TFMs) have recently emerged as the state-of-the-art for predictive tasks on structured data, without requiring bespoke kernels or hyperparameter tuning. We evaluate two TFMs, TabPFN and TabICL, as out-of-the-box surrogate models on two standard materials science benchmarks, comparing against a large field of baselines encompassing Gaussian processes, random forests, and CatBoost. Our findings indicate that TFMs are highly competitive surrogate models across the two benchmarks, without task-specific training for active learning or GP kernel selection. Our results motivate developing TFMs further as surrogates for active learning in materials discovery.

Figures & tables

Appendix figures & tables14 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Foundation-Model Surrogates Enable Data-Efficient Active Learning for Materials Discovery

    Mar 13, 2026Jeffrey Hu, Rongzhi Dong, Ying Feng +2Active LearningData-Efficient Learning

  2. Surrogate-Gated Generation and Foundation-Model Embeddings for Bayesian Materials Design

    Jun 26, 2026Sk Md Ahnaf Akif Alvi, Jan Janssen, Danny Perez +2Surrogate-Assisted OptimizationBayesian Optimization

  3. Transfer Learning Architectures for Scalable Multi-Fidelity Bayesian Optimization

    Jul 26, 2026Jaewook Lee, Ethan Errington, Christian D. Lorenz +1Multi-Fidelity OptimizationSurrogate Modeling