# Structured Latent Space Modeling over Multi-Scale Temporal Patches for Multivariate Time Series Forecasting

arXiv: https://arxiv.org/abs/2607.19404 · HTML: https://mlsanity.com/paper/2607.19404/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 51 references that are in the map.
- First posted on arXiv 7 Jul 2026.

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://mlsanity.com/paper/2012.07436/) (2020)
- [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://mlsanity.com/paper/2211.14730/) (2022)
- [Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting](https://mlsanity.com/paper/2106.13008/) (2021)
- [An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling](https://mlsanity.com/paper/1803.01271/) (2018)
- [Are Transformers Effective for Time Series Forecasting?](https://mlsanity.com/paper/2205.13504/) (2022)
- [FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting](https://mlsanity.com/paper/2201.12740/) (2022)
- [TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis](https://mlsanity.com/paper/2210.02186/) (2022)
- [iTransformer: Inverted Transformers Are Effective for Time Series Forecasting](https://mlsanity.com/paper/2310.06625/) (2023)
- [Transformers in Time Series: A Survey](https://mlsanity.com/paper/2202.07125/) (2022)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting](https://mlsanity.com/paper/2405.14616/) (2024)
- [Long-term Forecasting with TiDE: Time-series Dense Encoder](https://mlsanity.com/paper/2304.08424/) (2023)
- [Deep Time Series Models: A Comprehensive Survey and Benchmark](https://mlsanity.com/paper/2407.13278/) (2024)
- [A Riemannian Network for SPD Matrix Learning](https://mlsanity.com/paper/1608.04233/) (2016)
- [Is Mamba Effective for Time Series Forecasting?](https://mlsanity.com/paper/2403.11144/) (2024)
- [Dish-TS: A General Paradigm for Alleviating Distribution Shift in Time Series Forecasting](https://mlsanity.com/paper/2302.14829/) (2023)
- [TFDNet: Time-Frequency Enhanced Decomposed Network for Long-term Time Series Forecasting](https://mlsanity.com/paper/2308.13386/) (2023)
- [TopoCL: Topological Contrastive Learning for Time Series](https://mlsanity.com/paper/2502.02924/) (2025)
- [ManifoldFormer: Geometric Deep Learning for Neural Dynamics on Riemannian Manifolds](https://mlsanity.com/paper/2511.16828/) (2025)
- [DisenTS: Disentangled Channel Evolving Pattern Modeling for Multivariate Time Series Forecasting](https://mlsanity.com/paper/2410.22981/) (2024)
- [Time Series Forecasting via Direct Per-Step Probability Distribution Modeling](https://mlsanity.com/paper/2511.23260/) (2025)
- [ReCast: Reliability-aware Codebook Assisted Lightweight Time Series Forecasting](https://mlsanity.com/paper/2511.11991/) (2025)
- [SPDM: Geometry-Modulated State Space Modeling with Manifold Constraints for Time Series Forecasting](https://mlsanity.com/paper/2606.09917/) (2026)

## Abstract (arXiv)

Existing patching and multi-scale methods advance multivariate time series forecasting but treat learned representations as transient byproducts of prediction, lacking explicit mechanisms that enforce structural consistency across temporal scales. We propose M2Patch, a CNN-based architecture that organizes channel-independent observations into a structured latent space via two complementary differentiable penalties. Multi-scale patching decomposes the input into overlapping temporal granularities, depthwise separable CNN blocks with progressively growing dilation extracts scale-specific features at linear complexity, and per-scale learned projections compress these features into a compact latent representation. An intra-scale smoothness penalty enforces temporal continuity between adjacent patches, while an inter-scale alignment penalty restores cross-granularity interaction through learnable cross-scale mappings, so that all scales encode mutually consistent representations of the underlying dynamics. Extensive experiments on ten real-world benchmark datasets demonstrate that M2Patch significantly outperforms state-of-the-art baselines. Further analyses establish M2Patch as a structure-aware recognizer: it recovers channel functional groupings and remains robust under patch-level input corruption, confirming that the structured latent space captures the data's intrinsic dynamics.