# TimePro: Efficient Multivariate Long-term Time Series Forecasting with Variable- and Time-Aware Hyper-state

arXiv: https://arxiv.org/abs/2505.20774 · HTML: https://mlsanity.com/paper/2505.20774/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 32 references that are in the map.
- First posted on arXiv 27 May 2025.

## Builds on (most cited first, 23 of 23)

- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Efficiently Modeling Long Sequences with Structured State Spaces](https://mlsanity.com/paper/2111.00396/) (2021)
- [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)
- [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)
- [Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks](https://mlsanity.com/paper/1703.07015/) (2017)
- [Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers](https://mlsanity.com/paper/2110.13985/) (2021)
- [Non-stationary Transformers: Exploring the Stationarity in Time Series Forecasting](https://mlsanity.com/paper/2205.14415/) (2022)
- [Diagonal State Spaces are as Effective as Structured State Spaces](https://mlsanity.com/paper/2203.14343/) (2022)
- [TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting](https://mlsanity.com/paper/2405.14616/) (2024)
- [SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction](https://mlsanity.com/paper/2106.09305/) (2021)
- [Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structures](https://mlsanity.com/paper/2207.01186/) (2022)
- [Is Mamba Effective for Time Series Forecasting?](https://mlsanity.com/paper/2403.11144/) (2024)
- [SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion](https://mlsanity.com/paper/2404.14197/) (2024)
- [TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting](https://mlsanity.com/paper/2403.09898/) (2024)
- [Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications](https://mlsanity.com/paper/2401.06197/) (2024)
- [Recurrent Neural Networks for Time Series Forecasting](https://mlsanity.com/paper/1901.00069/) (2019)
- [Bi-Mamba+: Bidirectional Mamba for Time Series Forecasting](https://mlsanity.com/paper/2404.15772/) (2024)
- [Channel-Aware Low-Rank Adaptation in Time Series Forecasting](https://mlsanity.com/paper/2407.17246/) (2024)
- [A Mamba Foundation Model for Time Series Forecasting](https://mlsanity.com/paper/2411.02941/) (2024)
- [Linearly-evolved Transformer for Pan-sharpening](https://mlsanity.com/paper/2404.12804/) (2024)
- [Contextualizing MLP-Mixers Spatiotemporally for Urban Data Forecast at Scale](https://mlsanity.com/paper/2307.01482/) (2023)

## Built on by (most cited first, 16 of 16)

- [AlphaCast: A Human Wisdom-LLM Intelligence Co-Reasoning Framework for Interactive Time Series Forecasting](https://mlsanity.com/paper/2511.08947/) (2025)
- [Enhancing Multivariate Time Series Forecasting with Global Temporal Retrieval](https://mlsanity.com/paper/2602.10847/) (2026)
- [TimeExpert: Boosting Long Time Series Forecasting with Temporal Mix of Experts](https://mlsanity.com/paper/2509.23145/) (2025)
- [vLinear: A Powerful Linear Model for Multivariate Time Series Forecasting](https://mlsanity.com/paper/2601.13768/) (2026)
- [DMamba: Decomposition-enhanced Mamba for Time Series Forecasting](https://mlsanity.com/paper/2602.09081/) (2026)
- [DualWeaver: Synergistic Feature Weaving Surrogates for Multivariate Forecasting with Univariate Time Series Foundation Models](https://mlsanity.com/paper/2602.22066/) (2026)
- [What If We Let Forecasting Forget? A Sparse Bottleneck for Cross-Variable Dependencies](https://mlsanity.com/paper/2605.08289/) (2026)
- [Ada-MoGE: Adaptive Mixture of Gaussian Expert Model for Time Series Forecasting](https://mlsanity.com/paper/2512.02061/) (2025)
- [ACFormer: Mitigating Non-linearity with Auto Convolutional Encoder for Time Series Forecasting](https://mlsanity.com/paper/2601.20611/) (2026)
- [Permutation-Equivariant 2D State Space Models: Theory and Canonical Architecture for Multivariate Time Series](https://mlsanity.com/paper/2603.08753/) (2026)
- [MICA: Multivariate Infini Compressive Attention for Time Series Forecasting](https://mlsanity.com/paper/2604.06473/) (2026)
- [Generalizing Multi-Scale Time-Series Modeling with a Single Operator](https://mlsanity.com/paper/2605.31129/) (2026)
- [VLBM: Variational Latent Basis Modeling for OOD Robust Multivariate Time Series Forecasting](https://mlsanity.com/paper/2606.02138/) (2026)
- [TSSM: Triaxial State Space Model for Global Station Weather Forecasting with Temporal-Variable-Historical Modeling](https://mlsanity.com/paper/2607.13101/) (2026)
- [CARNet Cycle-Conditioned Core Aggregation and Redistribution for Multivariate Time Series Forecasting](https://mlsanity.com/paper/2607.21681/) (2026)
- [UniGIO: Unified Generative Global In-situ Weather Modeling from Spatiotemporal Incomplete Observations](https://mlsanity.com/paper/2609.22217/) (2026)

## Abstract (arXiv)

In long-term time series forecasting, different variables often influence the target variable over distinct time intervals, a challenge known as the multi-delay issue. Traditional models typically process all variables or time points uniformly, which limits their ability to capture complex variable relationships and obtain non-trivial time representations. To address this issue, we propose TimePro, an innovative Mamba-based model that constructs variate- and time-aware hyper-states. Unlike conventional approaches that merely transfer plain states across variable or time dimensions, TimePro preserves the fine-grained temporal features of each variate token and adaptively selects the focused time points to tune the plain state. The reconstructed hyper-state can perceive both variable relationships and salient temporal information, which helps the model make accurate forecasting. In experiments, TimePro performs competitively on eight real-world long-term forecasting benchmarks with satisfactory linear complexity. Code is available at https://github.com/xwmaxwma/TimePro.