SVTime: Small Time Series Forecasting Models Informed by "Physics" of Large Vision Model Forecasters
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912679025704960 |
|---|---|
| author | Shen, ChengAo Zhao, Ziming Tong, Hanghang Song, Dongjin Luo, Dongsheng Wen, Qingsong Ni, Jingchao |
| author_facet | Shen, ChengAo Zhao, Ziming Tong, Hanghang Song, Dongjin Luo, Dongsheng Wen, Qingsong Ni, Jingchao |
| contents | Time series AI is crucial for analyzing dynamic web content, driving a surge of pre-trained large models known for their strong knowledge encoding and transfer capabilities across diverse tasks. However, given their energy-intensive training, inference, and hardware demands, using large models as a one-fits-all solution raises serious concerns about carbon footprint and sustainability. For a specific task, a compact yet specialized, high-performing model may be more practical and affordable, especially for resource-constrained users such as small businesses. This motivates the question: Can we build cost-effective lightweight models with large-model-like performance on core tasks such as forecasting? This paper addresses this question by introducing SVTime, a novel Small model inspired by large Vision model (LVM) forecasters for long-term Time series forecasting (LTSF). Recently, LVMs have been shown as powerful tools for LTSF. We identify a set of key inductive biases of LVM forecasters -- analogous to the "physics" governing their behaviors in LTSF -- and design small models that encode these biases through meticulously crafted linear layers and constraint functions. Across 21 baselines spanning lightweight, complex, and pre-trained large models on 8 benchmark datasets, SVTime outperforms state-of-the-art (SOTA) lightweight models and rivals large models with 10^3 fewer parameters than LVMs, while enabling efficient training and inference in low-resource settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_09780 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SVTime: Small Time Series Forecasting Models Informed by "Physics" of Large Vision Model Forecasters Shen, ChengAo Zhao, Ziming Tong, Hanghang Song, Dongjin Luo, Dongsheng Wen, Qingsong Ni, Jingchao Machine Learning Artificial Intelligence Time series AI is crucial for analyzing dynamic web content, driving a surge of pre-trained large models known for their strong knowledge encoding and transfer capabilities across diverse tasks. However, given their energy-intensive training, inference, and hardware demands, using large models as a one-fits-all solution raises serious concerns about carbon footprint and sustainability. For a specific task, a compact yet specialized, high-performing model may be more practical and affordable, especially for resource-constrained users such as small businesses. This motivates the question: Can we build cost-effective lightweight models with large-model-like performance on core tasks such as forecasting? This paper addresses this question by introducing SVTime, a novel Small model inspired by large Vision model (LVM) forecasters for long-term Time series forecasting (LTSF). Recently, LVMs have been shown as powerful tools for LTSF. We identify a set of key inductive biases of LVM forecasters -- analogous to the "physics" governing their behaviors in LTSF -- and design small models that encode these biases through meticulously crafted linear layers and constraint functions. Across 21 baselines spanning lightweight, complex, and pre-trained large models on 8 benchmark datasets, SVTime outperforms state-of-the-art (SOTA) lightweight models and rivals large models with 10^3 fewer parameters than LVMs, while enabling efficient training and inference in low-resource settings. |
| title | SVTime: Small Time Series Forecasting Models Informed by "Physics" of Large Vision Model Forecasters |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2510.09780 |