Research map / Time series / Long-term forecasting
SEED: Spectral Entropy-Guided Evaluation of Spatial-Temporal Dependencies for Multivariate Time Series Forecasting
Feng Xiong, Zongxia Xie, Yanru Sun, Haoyu Wang, Jianhong Lin
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 the Context Spatial Extractor (local temporal windows) (ablation) | SEED · ECL | MSE: 0.178 -> 0.156 | Improved |
| add Temporal Attention branch (ablation) | SEED · ECL | MSE: 0.159 -> 0.156 | Improved |
| preserve negative correlations with a signed graph instead of plain softmax (ablation) | SEED · ECL | MSE: 0.177 -> 0.156 | Improved |
| use tanh-based instead of softmax-based signed graph (ablation) | SEED · ECL | MSE: 0.162 -> 0.156 | Improved |
| use spectral entropy fusion weights instead of learnable weights (ablation) | SEED · Solar-Energy | MSE: 0.231 -> 0.218 | Improved |
| use correct pairing of Temporal Attention and Context Spatial Extractor instead of cross-use (ablation) | SEED · ECL | MSE: 0.160 -> 0.156 | Improved |
| use local-window spatial context instead of same-moment relations only (ablation) | SEED · ECL | MSE: 0.167 -> 0.156 | Improved |
| use local-window spatial context instead of fully connected patch graph (ablation) | SEED · ECL | MSE: 0.164 -> 0.156 | Improved |
| apply SEED on datasets with many variables (PEMS03, PEMS07) | SEED · PEMS03, PEMS07 | MSE: 0.084 vs 0.087 (SOFTS) on PEMS03 | Mixed |
| SEED versus TQNet on Solar-Energy | SEED · Solar-Energy | MSE: 0.218 vs 0.198 (TQNet) | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 2 of 2 shared results
- SOFTS — SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion (NeurIPS 2024) · better on 2 of 2 shared results
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
- TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting (ICML 2025 · similarity 0.88)
- Introducing Spectral Attention for Long-Range Dependency in Time Series Forecasting (NeurIPS 2024 · similarity 0.88)
- Addressing Spatial-Temporal Heterogeneity: General Mixed Time Series Analysis via Latent Continuity Recovery and Alignment (NeurIPS 2024 · similarity 0.88)
- Causal Spatio-Temporal Prediction: An Effective and Efficient Multi-Modal Approach (NeurIPS 2025 · similarity 0.87)
- On the Integration of Spatial-Temporal Knowledge: A Lightweight Approach to Atmospheric Time Series Forecasting (NeurIPS 2025 · similarity 0.87)
- TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting (NeurIPS 2025 · similarity 0.86)
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.
