Research map / Time series / Long-term forecasting
SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement
Xiangfei Qiu, Xvyuan Liu, Tianen Shen, Xingjian Wu, Hanyin Cheng, Bin Yang, Jilin Hu
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 Mixture-of-Experts to patch embedding (ablation) | SEER · ETTh2 | MSE: 0.365 -> 0.356 | Improved |
| add Mixture-of-Experts to patch embedding (ablation) | SEER · Solar | MSE: 0.225 -> 0.221 | Improved |
| use augmented series embedding instead of simple MLP embedding (ablation) | SEER · ETTh2 | MSE: 0.379 -> 0.356 | Improved |
| use augmented series embedding instead of simple MLP embedding (ablation) | SEER · Solar | MSE: 0.240 -> 0.221 | Improved |
| add Automated Token Filter (ablation) | SEER · ETTh2 | MSE: 0.367 -> 0.356 | Improved |
| add Automated Token Filter (ablation) | SEER · Solar | MSE: 0.225 -> 0.221 | No clear effect |
| add Feature Reduction in flatten head (ablation) | SEER · ETTh2 | MSE: 0.372 -> 0.356 | Improved |
| add Feature Reduction in flatten head (ablation) | SEER · ETTm2 | MSE: 0.282 -> 0.276 | Improved |
| set Token Filter score threshold to non-zero (ablation) | SEER · most datasets | MSE | Improved |
| set Token Filter score threshold to non-zero (ablation) | SEER · Solar | MSE | Hurt |
| use patch length 16 or 24 rather than very small or large (ablation) | SEER · most datasets | MSE | Improved |
| use 8 experts in MoE (ablation) | SEER · most datasets | MSE | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- xPatch — xPatch: Dual-Stream Time Series Forecasting with Exponential Seasonal-Trend Decomposition (AAAI 2025) · better on 3 of 3 shared results
- Amplifier — Amplifier: Bringing Attention to Neglected Low-Energy Components in Time Series Forecasting (AAAI 2025) · better on 2 of 2 shared results
- iTransformer — iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 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
- Unlocking the Power of Patch: Patch-Based MLP for Long-Term Time Series Forecasting (AAAI 2025 · similarity 0.91)
- DeformableTST: Transformer for Time Series Forecasting without Over-reliance on Patching (NeurIPS 2024 · similarity 0.89)
- Sequence Complementor: Complementing Transformers for Time Series Forecasting with Learnable Sequences (AAAI 2025 · similarity 0.88)
- CARD: Channel Aligned Robust Blend Transformer for Time Series Forecasting (ICLR 2024 · similarity 0.88)
- A Closer Look at Transformers for Time Series Forecasting: Understanding Why They Work and Where They Struggle (ICML 2025 · similarity 0.88)
- iTransformer: Inverted Transformers Are Effective for Time Series Forecasting (ICLR 2024 · similarity 0.88)
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.
