Research map / Time series / Long-term forecasting
Beyond Point Predictions: Manifold Expansion and Dual Alignment for Robust Time Series Distillation
Junyao Hong, Zesheng Lai, Xinyi Xiao, Suyang Zhou, Aodong Shen, Youyong Kong
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 DSD distillation from PatchTST | LMP-Net · ETTh1 | MSE: 0.452 -> 0.430 | Improved |
| add DSD distillation | LMP-Net · Weather | MSE: 0.254 -> 0.246 | Improved |
| add DSD distillation | LMP-Net · Traffic | MSE: 0.490 -> 0.479 | Improved |
| add manifold expansion (Expand-Evolve-Contract) (ablation) | DSD LMP-Net · Traffic | MSE: 0.615 -> 0.479 | Improved |
| add manifold expansion (ablation) | DSD LMP-Net · Electricity | MSE: 0.211 -> 0.179 | Improved |
| add manifold expansion (ablation) | DSD LMP-Net · ETTh1 | MSE: 0.447 -> 0.430 | Improved |
| add dual manifold alignment (ablation) | DSD LMP-Net · Electricity | MSE: 0.183 -> 0.179 | Mixed |
| add dual manifold alignment (ablation) | DSD LMP-Net · Traffic | MSE: 0.491 -> 0.479 | No clear effect |
| add regime-aware prediction distillation (PredKD) (ablation) | DSD LMP-Net · Traffic | MSE: 0.491 -> 0.479 | Improved |
| add PredKD (ablation) | DSD LMP-Net · ETTh1 | MSE: 0.446 -> 0.430 | Improved |
| distill from Crossformer teacher | LMP-Net · ETTh1 | MSE: 0.452 -> 0.444 | Improved |
| distill from Pathformer teacher | LMP-Net · Electricity | MSE: 0.182 -> 0.182 | No clear effect |
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.
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024) · better on 6 of 8 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
- From Teacher Pathways to Invariant Manifolds: Consensus Subspace Distillation for TSFMs (ICML 2026 · similarity 0.88)
- Progressively Knowledge Distillation via Re-parameterizing Diffusion Reverse Process (AAAI 2024 · similarity 0.86)
- Reinforced Cross-Domain Knowledge Distillation on Time Series Data (NeurIPS 2024 · similarity 0.86)
- Sundial: A Family of Highly Capable Time Series Foundation Models (ICML 2025 · similarity 0.85)
- Gap Preserving Distillation by Building Bidirectional Mappings with A Dynamic Teacher (ICLR 2025 · similarity 0.85)
- TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting (ICML 2025 · similarity 0.85)
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.
