Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.28253 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910109939007488 |
|---|---|
| author | Xu, Xianyong Zuo, Yuanjun Huang, Zhihong Qin, Yihan Xu, Haoxian Du, Leilei Wang, Haotian |
| author_facet | Xu, Xianyong Zuo, Yuanjun Huang, Zhihong Qin, Yihan Xu, Haoxian Du, Leilei Wang, Haotian |
| contents | Time series forecasting is vital across many domains, yet existing models struggle with fixed-length inputs and inadequate multi-scale modeling. We propose MR-CDM, a framework combining hierarchical multi-resolution trend decomposition, an adaptive embedding mechanism for variable-length inputs, and a multi-scale conditional diffusion process. Evaluations on four real-world datasets demonstrate that MR-CDM significantly outperforms state-of-the-art baselines (e.g., CSDI, Informer), reducing MAE and RMSE by approximately 6-10 to a certain degree. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_28253 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations Xu, Xianyong Zuo, Yuanjun Huang, Zhihong Qin, Yihan Xu, Haoxian Du, Leilei Wang, Haotian Machine Learning Artificial Intelligence Time series forecasting is vital across many domains, yet existing models struggle with fixed-length inputs and inadequate multi-scale modeling. We propose MR-CDM, a framework combining hierarchical multi-resolution trend decomposition, an adaptive embedding mechanism for variable-length inputs, and a multi-scale conditional diffusion process. Evaluations on four real-world datasets demonstrate that MR-CDM significantly outperforms state-of-the-art baselines (e.g., CSDI, Informer), reducing MAE and RMSE by approximately 6-10 to a certain degree. |
| title | MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.28253 |