Research map / Time series / Causal and generative time series
TEN-DM: Topology-Enhanced Diffusion Model for Spatio-Temporal Event Prediction
Yuxin Liu, Kaiming Wang, Chenguang Yang, Yulia Gel, Yuzhou Chen
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 temporal topological learning (TTL) (ablation) | TEN-DM · Theft | temporal RMSE: 0.425 -> 0.363 | Improved |
| add temporal topological learning (TTL) (ablation) | TEN-DM · COVID-19 | temporal RMSE: 4.60% increase when removed | Improved |
| add graph learning (GNN pre-training) (ablation) | TEN-DM · Theft | temporal RMSE: 0.391 -> 0.363 | Improved |
| add TQ-SA (ablation) | TEN-DM · COVID-19 | spatial Euclidean distance: 1.28% increase when removed | Improved |
| add graph learning (ablation) | TEN-DM · JPN Earthquake | spatial Euclidean distance: 0.24% increase when removed | Improved |
| add TQ-SA (ablation) | TEN-DM · Theft | temporal RMSE: 0.374 -> 0.363 | Improved |
| use zigzag persistence instead of ordinary persistent homology (ablation) | TEN-DM · 311 Service | temporal RMSE: 0.800 -> 0.793 | Improved |
| double patch size (ablation) | TEN-DM · 311 Service | temporal RMSE: 0.7920 -> 0.8040 | No clear effect |
| change spatial resolution to 64x64 tiles (ablation) | TEN-DM · 311 Service | temporal RMSE: 0.7920 -> 0.8082 | No clear effect |
| reverse zigzag direction (ablation) | TEN-DM · 311 Service | spatial Euclidean distance: 0.0547 -> 0.0546 | No clear effect |
| add extra adjacency matrix and trainable aggregation weights (ablation) | TEN-DM · 311 Service | temporal RMSE: 0.7920 -> 0.5978 | Mixed |
| remove duplicated events (ablation) | TEN-DM · 311 Service | temporal RMSE: 0.7920±0.026 -> 0.7920±0.013 | No clear effect |
Closest papers by meaning
- Interacting Diffusion Processes for Event Sequence Forecasting (ICML 2024 · similarity 0.88)
- Seahorse: A Unified Benchmarking Framework for Spatiotemporal Event Modeling (NeurIPS 2026 · similarity 0.87)
- State Space Models on Temporal Graphs: A First-Principles Study (NeurIPS 2024 · similarity 0.86)
- STD-PLM: Understanding Both Spatial and Temporal Properties of Spatial-Temporal Data with PLM (AAAI 2025 · similarity 0.86)
- EasyTPP: Towards Open Benchmarking Temporal Point Processes (ICLR 2024 · similarity 0.86)
- Learning Spatiotemporal Dynamical Systems from Point Process Observations (ICLR 2025 · 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.
