Research map / Time series / Long-term forecasting
STrans: Spontaneous Architecture Evolution for Adaptive Time Series Forecasting
Haoyi Jia
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 |
|---|---|---|---|
| search architecture per dataset with STrans instead of fixed iTransformer | STrans · Weather | improvement over iTransformer: 8.5% | Improved |
| search architecture per dataset with STrans instead of fixed iTransformer | STrans · Solar-Energy | improvement over iTransformer: 3.0% | Improved |
| search architecture per dataset with STrans instead of fixed iTransformer | STrans · Exchange | improvement over iTransformer: 5.6% | Improved |
| increase model dimension from 128 to 256 (ablation) | STrans · ETTh1 | MSE: 0.385 -> 0.380 | Improved |
| increase model dimension from 128 to 256 (ablation) | STrans · Exchange | MSE: 0.085 -> 0.082 | Improved |
| increase model dimension from 256 to 512 (ablation) | STrans · ETTh1 and Exchange | MSE: 0.380 -> 0.381 (ETTh1); 0.082 -> 0.082 (Exchange) | No clear effect |
| longer prediction horizon from 96 to 720 (searched architecture fixed) (ablation) | STrans · Exchange | MSE increase: 908.5% MSE increase (ETTh1: 27.0%) | Hurt |
| use searched architecture instead of manually designed baselines on ETTh2 and ETTm2 long horizons | STrans · ETTm2 / ETTh2 (horizon 96-720) | MSE: ETTm2 96: 0.179 vs PatchTST 0.175 | Mixed |
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 6 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
- 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)
- SAMformer: Unlocking the Potential of Transformers in Time Series Forecasting with Sharpness-Aware Minimization and Channel-Wise Attention (ICML 2024 · similarity 0.87)
- TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting (ICLR 2024 · similarity 0.86)
- Slimming the Fat-Tail: Morphing-Flow for Adaptive Time Series Modeling (ICML 2025 · similarity 0.86)
- Are Self-Attentions Effective for Time Series Forecasting? (NeurIPS 2024 · 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.
