Research map / Time series / Causal and generative time series
TimeSeed: Effective Time Series Forecasting with Sparse Endogenous Variables
Zhaowang Wu, Kaixin Deng, Hua Yan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Time Domain Aggregator (ablation) | TimeSeed · ETTh2 | MSE: 0.376 -> 0.272 | Improved |
| add Frequency Domain Aggregator (ablation) | TimeSeed · ETTh1 | MSE: 0.118 -> 0.096 | Improved |
| add high-energy frequency aggregation (AggHigh) (ablation) | TimeSeed · ETTh1 | MSE: 0.112 -> 0.096 | Improved |
| add Adaptive Scale Reconstructor (ablation) | TimeSeed · ETTm2 | MSE: 0.176 -> 0.149 | Improved |
| add Adaptive Scale Reconstructor (ablation) | TimeSeed · ETTh1 | MSE: 0.099 -> 0.096 | Improved |
| use two-stage reconstruction instead of direct forecasting (ablation) | TimeSeed · ETTh2, ETTm2 | MSE and MAE: average improvements of 27.4% and 14.2% in MSE and MAE | Improved |
| use two-stage reconstruction instead of direct forecasting (ablation) | TimeSeed · ETTm1 | MSE: 0.074 -> 0.068 | Mixed |
| increase sparsity ratio (more endogenous observations) from 4% to 50% (ablation) | TimeSeed · ETTh2 | MSE: 0.272 -> 0.227 | Improved |
| increase sparsity ratio (more endogenous observations) from 4% to 50% (ablation) | TimeSeed · ETTm2 | MSE: 0.149 -> 0.121 | Improved |
| increase number of high-energy components beyond about 8 (ablation) | TimeSeed · sensitivity analysis | MSE | No clear effect |
| increase number of upsampling resolutions from 1 to 2 (ablation) | TimeSeed ASR · ETT datasets | MSE | Improved |
| add imputation stage (PatchTST) before forecasting instead of TimeSeed | PatchTST/PatchTST pipeline · ETTh1 | MSE: 0.096 -> 0.153 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TimeXer — TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables (NeurIPS 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables (NeurIPS 2024 · similarity 0.91)
- TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting (ICML 2025 · similarity 0.88)
- Enhancing Foundation Models for Time Series Forecasting via Wavelet-based Tokenization (ICML 2025 · similarity 0.88)
- SparseTSF: Modeling Long-term Time Series Forecasting with *1k* Parameters (ICML 2024 · similarity 0.87)
- TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting (NeurIPS 2025 · similarity 0.87)
- Sequence Complementor: Complementing Transformers for Time Series Forecasting with Learnable Sequences (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
