This repository contains the official Hugging Face / PyTorch implementation of SymTime from our NeurIPS 2025 paper, Synthetic Series-Symbol Data Generation for Time Series Foundation Models.
Overview
SymTime is a lightweight time series foundation model designed to learn strong temporal representations from patch-based inputs. It is built for practical downstream use and supports easy loading through the Hugging Face AutoModel interface.
SymTime
The model takes a univariate time series, splits it into patches, and encodes the patch sequence with a transformer backbone. The repository includes the configuration, model definition, and a runnable example for inference.
Output: patch embeddings, optionally with a CLS token output
Backend: patch-based transformer encoder
Citation
If you find this code useful, please cite our paper.
@misc{wang2025syntheticseriessymboldatageneration,
title={Synthetic Series-Symbol Data Generation for Time Series Foundation Models},
author={Wenxuan Wang and Kai Wu and Yujian Betterest Li and Dan Wang and Xiaoyu Zhang},
year={2025},
eprint={2510.08445},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2510.08445},
}
Contact
If you have any questions or are interested in our view on the complex dynamics of time series, feel free to contact:
