Revisiting the Generic Transformer: Deconstructing a Strong Baseline for Time Series Foundation Models

TL;DR

This paper demonstrates that a standard patch Transformer architecture, with proper training and data strategies, can serve as a strong, scalable baseline for time series forecasting, achieving state-of-the-art zero-shot performance.

Contribution

It shows that a generic Transformer architecture, combined with systematic ablation studies, can outperform specialized models in time series forecasting.

Findings

Standard patch Transformer achieves state-of-the-art zero-shot performance.

Model scaling and data strategies are key to high performance.

Open-source model and detailed results promote reproducibility.

Abstract

The recent surge in Time Series Foundation Models has rapidly advanced the field, yet the heterogeneous training setups across studies make it difficult to attribute improvements to architectural innovations versus data engineering. In this work, we investigate the potential of a standard patch Transformer, demonstrating that this generic architecture achieves state-of-the-art zero-shot forecasting performance using a straightforward training protocol. We conduct a comprehensive ablation study that covers model scaling, data composition, and training techniques to isolate the essential ingredients for high performance. Our findings identify the key drivers of performance, while confirming that the generic architecture itself demonstrates excellent scalability. By strictly controlling these variables, we provide comprehensive empirical results on model scaling across multiple dimensions. We release our open-source model and detailed findings to establish a transparent, reproducible baseline for future research.