CURE: Context-driven Diffusion with Progressive Expansion for Single Domain Generalization in Time Series Classification

Yuhang Pei, Fanchun Meng, Wenrui Wu, Tao Ren, Yifan Wang, Wei Ju, Chao Zheng, Xiao Luo
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:97738-97755, 2026.

Abstract

This paper studies the problem of single domain generalization in time series classification, which aims to learn a generalized time series classification model using a single source domain. This problem is highly challenging due to unreliable supervision from domain scarcity. Although current approaches employ generative models for data augmentation, these synthesized samples often suffer from low diversity and intrinsic noise, leading to weak generalization ability. Towards this end, we propose a novel approach named Context-driven Diffusion with Progressive Expansion (CURE) for single domain generalization in time series classification. The core of our CURE is to generate semantic-aware and semantic-free contexts that jointly guide a conditional diffusion model for informative data expansion. In particular, our CURE first conducts representation disentanglement to extract semantic-aware and semantic-free representations from source data. To enhance generalizability through data synthesis, we not only retrieve reference time series trajectories with similar semantics for semantic-aware contexts, but also utilize adversarial strategies to learn semantic-free contexts. These contexts are integrated as joint conditions for a diffusion model, enabling diverse and reliable virtual data. To enhance expansion adaptability and stable optimization, we progressively update our semantic-free contexts via a memory bank and measure boundary properties for dynamic data filtering. Comprehensive experiments on benchmark datasets validate the effectiveness of CURE.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-pei26c, title = {{CURE}: Context-driven Diffusion with Progressive Expansion for Single Domain Generalization in Time Series Classification}, author = {Pei, Yuhang and Meng, Fanchun and Wu, Wenrui and Ren, Tao and Wang, Yifan and Ju, Wei and Zheng, Chao and Luo, Xiao}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {97738--97755}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/pei26c/pei26c.pdf}, url = {https://proceedings.mlr.press/v306/pei26c.html}, abstract = {This paper studies the problem of single domain generalization in time series classification, which aims to learn a generalized time series classification model using a single source domain. This problem is highly challenging due to unreliable supervision from domain scarcity. Although current approaches employ generative models for data augmentation, these synthesized samples often suffer from low diversity and intrinsic noise, leading to weak generalization ability. Towards this end, we propose a novel approach named Context-driven Diffusion with Progressive Expansion (CURE) for single domain generalization in time series classification. The core of our CURE is to generate semantic-aware and semantic-free contexts that jointly guide a conditional diffusion model for informative data expansion. In particular, our CURE first conducts representation disentanglement to extract semantic-aware and semantic-free representations from source data. To enhance generalizability through data synthesis, we not only retrieve reference time series trajectories with similar semantics for semantic-aware contexts, but also utilize adversarial strategies to learn semantic-free contexts. These contexts are integrated as joint conditions for a diffusion model, enabling diverse and reliable virtual data. To enhance expansion adaptability and stable optimization, we progressively update our semantic-free contexts via a memory bank and measure boundary properties for dynamic data filtering. Comprehensive experiments on benchmark datasets validate the effectiveness of CURE.} }
Endnote
%0 Conference Paper %T CURE: Context-driven Diffusion with Progressive Expansion for Single Domain Generalization in Time Series Classification %A Yuhang Pei %A Fanchun Meng %A Wenrui Wu %A Tao Ren %A Yifan Wang %A Wei Ju %A Chao Zheng %A Xiao Luo %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-pei26c %I PMLR %P 97738--97755 %U https://proceedings.mlr.press/v306/pei26c.html %V 306 %X This paper studies the problem of single domain generalization in time series classification, which aims to learn a generalized time series classification model using a single source domain. This problem is highly challenging due to unreliable supervision from domain scarcity. Although current approaches employ generative models for data augmentation, these synthesized samples often suffer from low diversity and intrinsic noise, leading to weak generalization ability. Towards this end, we propose a novel approach named Context-driven Diffusion with Progressive Expansion (CURE) for single domain generalization in time series classification. The core of our CURE is to generate semantic-aware and semantic-free contexts that jointly guide a conditional diffusion model for informative data expansion. In particular, our CURE first conducts representation disentanglement to extract semantic-aware and semantic-free representations from source data. To enhance generalizability through data synthesis, we not only retrieve reference time series trajectories with similar semantics for semantic-aware contexts, but also utilize adversarial strategies to learn semantic-free contexts. These contexts are integrated as joint conditions for a diffusion model, enabling diverse and reliable virtual data. To enhance expansion adaptability and stable optimization, we progressively update our semantic-free contexts via a memory bank and measure boundary properties for dynamic data filtering. Comprehensive experiments on benchmark datasets validate the effectiveness of CURE.
APA
Pei, Y., Meng, F., Wu, W., Ren, T., Wang, Y., Ju, W., Zheng, C. & Luo, X.. (2026). CURE: Context-driven Diffusion with Progressive Expansion for Single Domain Generalization in Time Series Classification. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:97738-97755 Available from https://proceedings.mlr.press/v306/pei26c.html.

Related Material