DySCo: Dynamic Semantic Compression for Effective Long-term Time Series Forecasting
This addresses the challenge of effectively capturing complex long-term dependencies in time series forecasting for domains like finance and energy, representing an incremental improvement over traditional methods.
The paper tackled the problem of irrelevant noise and computational redundancy in long-term time series forecasting by proposing the DySCo framework, which achieved significant enhancement in capturing long-term correlations with reduced computational cost as a universal plug-and-play module.
Time series forecasting (TSF) is critical across domains such as finance, meteorology, and energy. While extending the lookback window theoretically provides richer historical context, in practice, it often introduces irrelevant noise and computational redundancy, preventing models from effectively capturing complex long-term dependencies. To address these challenges, we propose a Dynamic Semantic Compression (DySCo) framework. Unlike traditional methods that rely on fixed heuristics, DySCo introduces an Entropy-Guided Dynamic Sampling (EGDS) mechanism to autonomously identify and retain high-entropy segments while compressing redundant trends. Furthermore, we incorporate a Hierarchical Frequency-Enhanced Decomposition (HFED) strategy to separate high-frequency anomalies from low-frequency patterns, ensuring that critical details are preserved during sparse sampling. Finally, a Cross-Scale Interaction Mixer(CSIM) is designed to dynamically fuse global contexts with local representations, replacing simple linear aggregation. Experimental results demonstrate that DySCo serves as a universal plug-and-play module, significantly enhancing the ability of mainstream models to capture long-term correlations with reduced computational cost.