探究 Chronos-2 与 TabPFN-TS 中简单目标-协变量关系
机器学习
2026-05-13 v1
摘要
Time Series Foundation Models (TSFMs) have recently achieved state-of-the-art performance, often outperforming supervised models in zero-shot settings. Recent TSFM architectures, such as Chronos-2 and TabPFN-TS, aim to integrate covariates. In this paper, we design controlled experiments based on simple target-covariate relationships to assess this integration capability. Our results show that TabPFN-TS captures these relationships more effectively than Chronos-2, especially for short horizons, suggesting that the strong benchmark performance of Chronos-2 does not automatically translate into optimal modeling of simple covariate-target dependencies.
引用
@article{arxiv.2605.12200,
title = {Investigating simple target-covariate relationships for Chronos-2 and TabPFN-TS},
author = {Gaspard Berthelier and Mariia Baranova and Andrei-Tiberiu Pantea and Etienne Le Naour and Adrien Petralia and Tahar Nabil and Themis Palpanas},
journal= {arXiv preprint arXiv:2605.12200},
year = {2026}
}