DCRA: Diffusion-Conditioned Representation Alignment for Robust Time-Series Learning

Abstract

Learning robust representations for time-series signals under noise and distribution shifts remains challenging, especially in clinical applications such as electroencephalogram (EEG) and electrocardiogram (ECG) analysis. We propose Diffusion-Conditioned Representation Alignment (DCRA), a training framework that repurposes the forward diffusion process as a structured corruption scheduler for representation learning. Different from conventional augmentation and consistency-based methods that rely on independently sampled perturbations, DCRA introduces a structured corruption trajectory via the diffusion forward process, which enables continuous and controlled representation evolution across noise levels. We introduce a feature-level consistency objective that aligns representations across noise levels while preserving class-discriminative structure. This mechanism promotes structure-preserving consistency, which enables smooth and semantically coherent feature trajectories in latent space. The proposed framework is encoder-agnostic and can be integrated with state space models and Transformer architectures. The seizure detection experiments on the CHB-MIT EEG dataset show that DCRA consistently improves performance under multiple noise conditions and achieves higher sensitivity at low false-positive rates. Analysis reveals that DCRA produces more balanced and structured representations compared to baseline and diffusion-only models. These findings highlight the benefit of combining structured corruption with representation alignment for robust time-series learning.

MiCLA: Mixture-to-Constituent Latent Alignment for Polyphonic Audio Representation Learning
Kohei Yamamoto, Kosuke Okusa
arXiv20 Oct 2026PreprintarXiv licenceFull textStatisticsarXiv
Class-Conditioned Diffusion Augmentation and Explainable In-Context Ensemble Learning for Imbalanced Maintenance Action Classification in Aircraft Systems
Begüm Erol
Mathematics4 Nov 2026Peer-reviewedCC BY 4.0General mathematicsCrossref
Dual-domain self-supervised feature alignment via spectral–spatial representation learning for deepfake anomaly detection
Yi He
PLOS One11 Sep 2026Peer-reviewedCC BY 4.0Full textBioinformatics & computational biology ProbabilityPLOS
Byzantine-Robust Federated Representation Learning
Leonardo F. Toso, James Anderson, Rafael Pinot +1
arXiv29 Sep 2026PreprintarXiv licenceFull textOptimisation & controlarXiv
Memory-Conditioned Diffusion Model for Generalized Langevin Dynamics
Minglei Yang, Sicheng He
arXiv23 Sep 2026PreprintarXiv licenceFull textMathematical physics StatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{xu2026dcra,
  title = {{DCRA: Diffusion-Conditioned Representation Alignment for Robust Time-Series Learning}},
  author = {Wenrui Xu and Anas Enanaa and Keshab K. Parhi},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.11997},
  eprint = {2609.11997},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.11997},
  note = {CC BY-NC-ND 4.0},
}

RIS

TY  - UNPB
TI  - DCRA: Diffusion-Conditioned Representation Alignment for Robust Time-Series Learning
AU  - Wenrui Xu
AU  - Anas Enanaa
AU  - Keshab K. Parhi
PY  - 2026
DA  - 2026/09/09
PB  - arXiv
DO  - 10.48550/arxiv.2609.11997
UR  - https://arxiv.org/abs/2609.11997
AB  - Learning robust representations for time-series signals under noise and distribution shifts remains challenging, especially in clinical applications such as electroencephalogram (EEG) and electrocardiogram (ECG) analysis. We propose Diffusion-Conditioned Representation Alignment (DCRA), a training framework that repurposes the forward diffusion process as a structured corruption scheduler for representation learning. Different from conventional augmentation and consistency-based methods that rely on independently sampled perturbations, DCRA introduces a structured corruption trajectory via the diffusion forward process, which enables continuous and controlled representation evolution across noise levels. We introduce a feature-level consistency objective that aligns representations across noise levels while preserving class-discriminative structure. This mechanism promotes structure-preserving consistency, which enables smooth and semantically coherent feature trajectories in latent space. The proposed framework is encoder-agnostic and can be integrated with state space models and Transformer architectures. The seizure detection experiments on the CHB-MIT EEG dataset show that DCRA consistently improves performance under multiple noise conditions and achieves higher sensitivity at low false-positive rates. Analysis reveals that DCRA produces more balanced and structured representations compared to baseline and diffusion-only models. These findings highlight the benefit of combining structured corruption with representation alignment for robust time-series learning.
ER  -

Abstract

Related

Cite