Research map / Time series / Long-term forecasting
GraFT: Infusing Pre-trained Transformers with Relational Structure for Time Series Forecasting
Yuqi Yuan, Xiong Luo, Qiaojuan Peng, Wenbing Zhao
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 all graph edge sets (full HPRG) vs none (ablation) | GraFT · ECL | MSE: 0.319 -> 0.113 | Improved |
| add all graph edge sets (full HPRG) vs none (ablation) | GraFT · ILI | MSE: 2.722 -> 1.243 | Improved |
| add similarity edges (ablation) | GraFT · ECL | MSE: 0.244 -> 0.113 | Improved |
| add lagged edges (ablation) | GraFT · ILI | MSE: 1.607 -> 1.243 | Improved |
| add adjacency edges (ablation) | GraFT · ECL | MSE: 0.155 -> 0.113 | Improved |
| add synchrony edges (ablation) | GraFT · ILI | MSE: 1.452 -> 1.243 | Improved |
| combine four edge types instead of adjacency edges alone (ablation) | GraFT · ILI | MSE: 2.049 -> 1.243 | Improved |
| use R-GCN instead of GCN encoder (ablation) | GraFT · ETTh1 | MSE: 0.393 -> 0.375 | Improved |
| use R-GCN instead of R-GAT (ablation) | GraFT · ETTh1 | MSE: 0.412 -> 0.375 | Improved |
| use R-GCN instead of MLP encoder (ablation) | GraFT · ETTm1 | MSE: 0.328 -> 0.296 | Improved |
| use GraFT instead of FSCA | FSCA · ECL | MSE: 0.159 -> 0.113 | Improved |
| use GraFT instead of FSCA | FSCA · ETTm1 | MSE: 0.342 -> 0.296 | Improved |
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.
- Time-LLM — Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (ICLR 2024) · better on 9 of 9 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
- HDT: Hierarchical Discrete Transformer for Multivariate Time Series Forecasting (AAAI 2025 · similarity 0.89)
- A Closer Look at Transformers for Time Series Forecasting: Understanding Why They Work and Where They Struggle (ICML 2025 · similarity 0.88)
- TStruct: Learning Shared Temporal Structures for Long-Term Time-Series Forecasting (NeurIPS 2026 · similarity 0.88)
- TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment (AAAI 2025 · similarity 0.88)
- Sequence Complementor: Complementing Transformers for Time Series Forecasting with Learnable Sequences (AAAI 2025 · similarity 0.88)
- Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (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.
