From Tables to Signals: Revealing Spectral Adaptivity in TabPFN

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zheng, Jianqiao, Gordon, Cameron, Ji, Yiping, Saratchandran, Hemanth, Lucey, Simon
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915634536775680
author Zheng, Jianqiao
Gordon, Cameron
Ji, Yiping
Saratchandran, Hemanth
Lucey, Simon
author_facet Zheng, Jianqiao
Gordon, Cameron
Ji, Yiping
Saratchandran, Hemanth
Lucey, Simon
contents Task-agnostic tabular foundation models such as TabPFN have achieved impressive performance on tabular learning tasks, yet the origins of their inductive biases remain poorly understood. In this work, we study TabPFN through the lens of signal reconstruction and provide the first frequency-based analysis of its in-context learning behavior. We show that TabPFN possesses a broader effective frequency capacity than standard ReLU-MLPs, even without hyperparameter tuning. Moreover, unlike MLPs whose spectra evolve primarily over training epochs, we find that TabPFN's spectral capacity adapts directly to the number of samples provided in-context, a phenomenon we term Spectral Adaptivity. We further demonstrate that positional encoding modulates TabPFN's frequency response, mirroring classical results in implicit neural representations. Finally, we show that these properties enable TabPFN to perform training-free and hyperparameter-free image denoising, illustrating its potential as a task-agnostic implicit model. Our analysis provides new insight into the structure and inductive biases of tabular foundation models and highlights their promise for broader signal reconstruction tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18278
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Tables to Signals: Revealing Spectral Adaptivity in TabPFN
Zheng, Jianqiao
Gordon, Cameron
Ji, Yiping
Saratchandran, Hemanth
Lucey, Simon
Machine Learning
Computer Vision and Pattern Recognition
Task-agnostic tabular foundation models such as TabPFN have achieved impressive performance on tabular learning tasks, yet the origins of their inductive biases remain poorly understood. In this work, we study TabPFN through the lens of signal reconstruction and provide the first frequency-based analysis of its in-context learning behavior. We show that TabPFN possesses a broader effective frequency capacity than standard ReLU-MLPs, even without hyperparameter tuning. Moreover, unlike MLPs whose spectra evolve primarily over training epochs, we find that TabPFN's spectral capacity adapts directly to the number of samples provided in-context, a phenomenon we term Spectral Adaptivity. We further demonstrate that positional encoding modulates TabPFN's frequency response, mirroring classical results in implicit neural representations. Finally, we show that these properties enable TabPFN to perform training-free and hyperparameter-free image denoising, illustrating its potential as a task-agnostic implicit model. Our analysis provides new insight into the structure and inductive biases of tabular foundation models and highlights their promise for broader signal reconstruction tasks.
title From Tables to Signals: Revealing Spectral Adaptivity in TabPFN
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.18278