Research map / Time series / Long-term forecasting
CrossGNN: Confronting Noisy Multivariate Time Series Via Cross Interaction Refinement
Qihe Huang, Lei Shen, Ruixin Zhang, Shouhong Ding, Binwu Wang, Zhengyang Zhou, Yang Wang
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 Cross-Scale GNN (ablation) | CrossGNN · Weather T=96 | MSE: 0.175 -> 0.159 | Improved |
| add Cross-Scale GNN (ablation) | CrossGNN · ETTm2 T=96 | MSE: 0.200 -> 0.176 | Improved |
| add adaptive multi-scale identifier (vs fixed pooling lengths) (ablation) | CrossGNN · Weather T=96 | MSE: 0.167 -> 0.159 | Improved |
| add adaptive multi-scale identifier (ablation) | CrossGNN · Traffic T=96 | MSE: 0.585 -> 0.570 | Improved |
| add heterogeneous connections (ablation) | CrossGNN · ETTm2 T=96 | MSE: 0.190 -> 0.176 | Improved |
| add Cross-Variable GNN (homogeneous and heterogeneous) (ablation) | CrossGNN · Weather T=96 | MSE: 0.174 -> 0.159 | Improved |
| add Cross-Variable GNN (ablation) | CrossGNN · Traffic T=96 | MSE: 0.588 -> 0.570 | Improved |
| use CrossGNN instead of best Transformer-based method | CrossGNN · 8 MTS datasets | MSE: 10.43% reduction | Improved |
| use CrossGNN instead of MTGNN | CrossGNN · 8 MTS datasets | MSE: 22.57% reduction | Improved |
| use CrossGNN instead of PatchTST | CrossGNN · Electricity T=96 | MSE: 0.159 -> 0.173 | Hurt |
| lower SNR from 100db to 0db | CrossGNN vs ETSformer, MTGNN · ETTm2 | MSE: 0.177 vs 0.191 and 0.205 | Improved |
| increase number of scales from 4 to 8 | CrossGNN · Weather and Traffic | MSE and MAE | No clear effect |
Later papers that compared with this method
2 later accepted papers report CrossGNN as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- CPiRi: Channel Permutation-Invariant Relational Interaction for Multivariate Time Series Forecasting (ICLR 2026 · better on 1 of 1 shared result)
- Rethinking Fourier Transform from A Basis Functions Perspective for Long-term Time Series Forecasting (NeurIPS 2024 · better on 2 of 2 shared results)
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.
- TimesNet — TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (ICLR 2023) · better on 7 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
- Multivariate Time Series Forecasting needs Cross Variable Loss (NeurIPS 2026 · similarity 0.88)
- TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data (AAAI 2025 · similarity 0.88)
- CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuning (AAAI 2025 · similarity 0.88)
- Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach (ICML 2024 · similarity 0.88)
- MSGNet: Learning Multi-Scale Inter-series Correlations for Multivariate Time Series Forecasting (AAAI 2024 · similarity 0.88)
- Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling (ICML 2024 · 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.
