# Rethinking Time Series Domain Generalization via Structure-Stratified Calibration

arXiv: https://arxiv.org/abs/2603.02756 · HTML: https://mlsanity.com/paper/2603.02756/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 48 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Builds on (most cited first, 25 of 25)

- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Deep CORAL: Correlation Alignment for Deep Domain Adaptation](https://mlsanity.com/paper/1607.01719/) (2016)
- [Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2002.08546/) (2020)
- [FDA: Fourier Domain Adaptation for Semantic Segmentation](https://mlsanity.com/paper/2004.05498/) (2020)
- [Non-stationary Transformers: Exploring the Stationarity in Time Series Forecasting](https://mlsanity.com/paper/2205.14415/) (2022)
- [FSDR: Frequency Space Domain Randomization for Domain Generalization](https://mlsanity.com/paper/2103.02370/) (2021)
- [Causality Inspired Representation Learning for Domain Generalization](https://mlsanity.com/paper/2203.14237/) (2022)
- [A Survey on Negative Transfer](https://mlsanity.com/paper/2009.00909/) (2020)
- [DIVA: Domain Invariant Variational Autoencoders](https://mlsanity.com/paper/1905.10427/) (2019)
- [UniTS: A Unified Multi-Task Time Series Model](https://mlsanity.com/paper/2403.00131/) (2024)
- [Domain Adaptation for Time Series Under Feature and Label Shifts](https://mlsanity.com/paper/2302.03133/) (2023)
- [Contrastive Learning for Unsupervised Domain Adaptation of Time Series](https://mlsanity.com/paper/2206.06243/) (2022)
- [Semantics, Distortion, and Style Matter: Towards Source-free UDA for Panoramic Segmentation](https://mlsanity.com/paper/2403.12505/) (2024)
- [Test-time Fourier Style Calibration for Domain Generalization](https://mlsanity.com/paper/2205.06427/) (2022)
- [Contrastive ACE: Domain Generalization Through Alignment of Causal Mechanisms](https://mlsanity.com/paper/2106.00925/) (2021)
- [Pareto Domain Adaptation](https://mlsanity.com/paper/2112.04137/) (2021)
- [Generalizable Sleep Staging via Multi-Level Domain Alignment](https://mlsanity.com/paper/2401.05363/) (2023)
- [Mitigating Noisy Correspondence by Geometrical Structure Consistency Learning](https://mlsanity.com/paper/2405.16996/) (2024)
- [Enhancing Evolving Domain Generalization through Dynamic Latent Representations](https://mlsanity.com/paper/2401.08464/) (2024)
- [Time-Series Domain Adaptation via Sparse Associative Structure Alignment: Learning Invariance and Variance](https://mlsanity.com/paper/2205.03554/) (2022)
- [Hierarchical Variational Auto-Encoding for Unsupervised Domain Generalization](https://mlsanity.com/paper/2101.09436/) (2021)
- [Deep Unsupervised Domain Adaptation for Time Series Classification: a Benchmark](https://mlsanity.com/paper/2312.09857/) (2023)
- [Geometric Understanding of Discriminability and Transferability for Visual Domain Adaptation](https://mlsanity.com/paper/2407.09524/) (2024)
- [Geometric Moment Alignment for Domain Adaptation via Siegel Embeddings](https://mlsanity.com/paper/2510.14666/) (2025)

## Abstract (arXiv)

For time series arising from latent dynamical systems, existing cross-domain generalization methods commonly assume that samples are comparably meaningful within a shared representation space. In real-world settings, however, different datasets often originate from structurally heterogeneous families of dynamical systems, leading to fundamentally distinct feature distributions. Under such circumstances, performing global alignment while neglecting structural differences is highly prone to establishing spurious correspondences and inducing negative transfer. From the new perspective of cross-domain structural correspondence failure, we revisit this problem and propose a structurally stratified calibration framework. This approach explicitly distinguishes structurally consistent samples and performs amplitude calibration exclusively within structurally compatible sample clusters, thereby effectively alleviating generalization failures caused by structural incompatibility. Notably, the proposed framework achieves substantial performance improvements through a concise and computationally efficient calibration strategy. Evaluations on 19 public datasets (100.3k samples) demonstrate that SSCF significantly outperforms strong baselines under the zero-shot setting. These results confirm that establishing structural consistency prior to alignment constitutes a more reliable and effective pathway for improving cross-domain generalization of time series governed by latent dynamical systems.