Research map / Time series / Long-term forecasting
Revitalizing Canonical Pre-Alignment for Irregular Multivariate Time Series Forecasting
Ziyu Zhou, Yiming Huang, Yanyun Wang, Yuankai Wu, James Kwok, Yuxuan Liang
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 canonical pre-alignment (ablation) | KAFNet · PhysioNet | MSE x1e-3: 6.21 -> 5.88 | Improved |
| add Pre-Convolution (ablation) | KAFNet · PhysioNet | MSE x1e-3: 6.42 -> 5.88 | Improved |
| add T-Norm (ablation) | KAFNet · PhysioNet | MSE x1e-3: 6.37 -> 5.88 | Improved |
| add Temporal Kernel Aggregation (ablation) | KAFNet · PhysioNet | MSE x1e-3: 6.95 -> 5.88 | Improved |
| add Frequency Linear Attention (ablation) | KAFNet · PhysioNet | MSE x1e-3: 6.26 -> 5.88 | Improved |
| add Frequency Linear Attention in place of softmax attention (ablation) | KAFNet · PhysioNet | MSE x1e-3: 6.08 -> 5.88 | Improved |
| add canonical pre-alignment (ablation) | KAFNet · MIMIC | MSE x1e-2: 1.69 -> 1.59 | Improved |
| add Pre-Convolution (ablation) | KAFNet · MIMIC | MSE x1e-2: 1.62 -> 1.59 | Improved |
| add T-Norm (ablation) | KAFNet · MIMIC | MSE x1e-2: 1.73 -> 1.59 | Improved |
| add Temporal Kernel Aggregation (ablation) | KAFNet · MIMIC | MSE x1e-2: 1.74 -> 1.59 | Improved |
| add Frequency Linear Attention (ablation) | KAFNet · MIMIC | MSE x1e-2: 1.79 -> 1.59 | Improved |
| add Temporal Kernel Aggregation (ablation) | KAFNet · Human Activity | MSE x1e-3: 4.21 -> 2.54 | 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.
- GraFITi* — GraFITi: Graphs for Forecasting Irregularly Sampled Time Series (AAAI 2024) · better on 8 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
- Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach (ICML 2024 · similarity 0.91)
- Time-IMM: A Dataset and Benchmark for Irregular Multimodal Multivariate Time Series (NeurIPS 2025 · similarity 0.88)
- TimeCHEAT: A Channel Harmony Strategy for Irregularly Sampled Multivariate Time Series Analysis (AAAI 2025 · similarity 0.87)
- HyperIMTS: Hypergraph Neural Network for Irregular Multivariate Time Series Forecasting (ICML 2025 · similarity 0.87)
- IMTS is Worth Time $\times$ Channel Patches: Visual Masked Autoencoders for Irregular Multivariate Time Series Prediction (ICML 2025 · similarity 0.87)
- Physiome-ODE: A Benchmark for Irregularly Sampled Multivariate Time-Series Forecasting Based on Biological ODEs (ICLR 2025 · similarity 0.87)
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.
