Research map / Graph learning / Knowledge and temporal graphs
ComS2T: A Complementary Spatiotemporal Learning System for Data-Adaptive Model Evolution
Zhengyang Zhou, Qihe Huang, Binwu Wang, Jianpeng Hou, Kuo Yang, Yuxuan Liang, Yu Zheng, 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 hippocampus/neocortex disentanglement (ablation) | ComS2T · SIP (temporal shift) | MAE: 54.48 -> 41.12 | Improved |
| add hippocampus/neocortex disentanglement (ablation) | ComS2T · Metr-LA (temporal shift) | MAE: 5.48 -> 4.38 | Improved |
| add hippocampus/neocortex disentanglement (ablation) | ComS2T · Temperature (temporal shift) | MAE: 6.91 -> 6.82 | Improved |
| add self-supervised prompt training (ablation) | ComS2T · SIP (temporal shift) | MAE: 45.73 -> 41.12 | Improved |
| add self-supervised prompt training (ablation) | ComS2T · KnowAir (temporal shift) | MAE: 36.46 -> 35.32 | Improved |
| add spatial-temporal prompts (ablation) | ComS2T · KnowAir (temporal shift) | MAE: 38.33 -> 35.32 | Improved |
| add spatial-temporal prompts (ablation) | ComS2T · SIP (node involvement) | MAE: 46.47 -> 39.88 | Improved |
| add test-time prompt update (ablation) | ComS2T · Temperature (temporal shift) | MAE: 7.27 -> 6.82 | Improved |
| add test-time prompt update (ablation) | ComS2T · SIP (temporal shift) | MAE: 45.12 -> 41.12 | Improved |
| add test-time prompt update (ablation) | ComS2T · KnowAir (temporal shift) | MAE: 35.14 -> 35.32 | Hurt |
| use ComS2T instead of PECPM | ComS2T · Metr-LA (node removal) | MAE: 6.80 -> 5.62 | Improved |
| use ComS2T instead of best baseline CaST | ComS2T · KnowAir (node involvement) | MAE: 33.42 -> 34.48 | Hurt |
Closest papers by meaning
- Get Rid of Isolation: A Continuous Multi-task Spatio-Temporal Learning Framework (NeurIPS 2024 · similarity 0.89)
- Diffusion Transformers as Open-World Spatiotemporal Foundation Models (NeurIPS 2025 · similarity 0.89)
- Drawing Informative Gradients from Sources: A One-stage Transfer Learning Framework for Cross-city Spatiotemporal Forecasting (AAAI 2025 · similarity 0.87)
- SynEVO: A neuro-inspired spatiotemporal evolutional framework for cross-domain adaptation (ICML 2025 · similarity 0.87)
- UniSTD: Towards Unified Spatio-Temporal Learning across Diverse Disciplines (CVPR 2025 · similarity 0.86)
- Spatio-Temporal Few-Shot Learning via Diffusive Neural Network Generation (ICLR 2024 · similarity 0.85)
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.
