Transformer-based tabular foundation models (TFMs) dominate small to medium tabular predictive benchmark tasks, yet their inference mechanisms remain largely unexplored. We present the first large-scale mechanistic study of layerwise dynamics in 6 state-of-the-art tabular in-context learning models. We explore how predictions emerge across depth, identify distinct stages of inference and reveal latent-space dynamics that differ from those of language models. Our findings indicate substantial depthwise redundancy across multiple models, suggesting iterative refinement with overlapping computations during inference stages. Guided by these insights, we design a proof-of-concept, looped single-layer model that uses only 20% of the original model's parameters while achieving comparable performance. The code is available at https://github.com/amirbalef/is_one_layer_enough.
@article{arxiv.2605.06510,
title = {Is One Layer Enough? Understanding Inference Dynamics in Tabular Foundation Models},
author = {Amir Rezaei Balef and Mykhailo Koshil and Katharina Eggensperger},
journal= {arXiv preprint arXiv:2605.06510},
year = {2026}
}
Comments
Accepted at the 43rd International Conference on Machine Learning (ICML 2026)