PULSE: Generative Phase Evolution for Non-Stationary Time Series Forecasting
About
Time series forecasting under non-stationarity faces a fundamental tension between capturing stable representations and adapting to distribution shifts. Existing methods implicitly rely on static historical assumptions, leading to a critical failure mode we term Phase Amnesia, where models become blind to the evolving global context. To resolve this, we formalize non-stationary dynamics through three physical hypotheses: wold decomposition, dynamical phase evolution, and heteroscedastic manifold generation. These principles inspire PULSE, a physics-informed, plug-and-play framework adopting a Disentangle--Evolve--Simulate design philosophy. Specifically, PULSE utilizes phase-anchored disentanglement to resolve optimization interference caused by dominant trends, employs a Phase Router to actively generate future trajectories, and introduces Statistic-Aware Mixup (SAM) to ensure robustness against out-of-distribution volatility. Empirically, PULSE enables a simple MLP backbone to achieve state-of-the-art or highly competitive performance across 12 real-world benchmarks. This validates that a correct physics-informed inductive bias is far more critical than raw architectural complexity for non-stationary forecasting. The code is available at: https://github.com/Gemost/PULSE.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Time Series Forecasting | ETTh1 | MSE0.409 | 836 | |
| Multivariate Forecasting | ETTh1 | MSE0.412 | 830 | |
| Multivariate Time-series Forecasting | ETTm1 | MSE0.363 | 686 | |
| Multivariate Time-series Forecasting | ETTm2 | MSE0.262 | 539 | |
| Time Series Forecasting | Weather | MSE0.219 | 497 | |
| Multivariate Time-series Forecasting | Weather | MSE0.239 | 409 | |
| Multivariate Time-series Forecasting | Traffic | MSE0.46 | 310 | |
| Time Series Forecasting | ETTm2 | MSE0.247 | 300 | |
| Time Series Forecasting | Electricity | MSE0.149 | 237 | |
| Multivariate Time-series Forecasting | ETTh2 | MSE0.361 | 198 |