Research map / Probabilistic and scientific modeling / Causal discovery
CausalX: A Unified and Causally-Interpretable Plug-and-Play Model for Multi-modal Spatio-Temporal Forecasting
Shiqi Zhang, Pan Mu, Hanting Yan, Yuchao Zhu, jinglin zhang, Cong Bai
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 |
|---|---|---|---|
| plug CausalX into MID | MID · ETH/UCY avg | ADE/FDE: 0.248/0.458 -> 0.238/0.420 | Improved |
| plug CausalX into EigenTrajectory | EigenTrajectory · ETH/UCY ZARA1 | ADE: 0.205 -> 0.213 | Hurt |
| plug CausalX into EigenTrajectory | EigenTrajectory · ETH/UCY avg | ADE: 0.224 -> 0.222 | Improved |
| plug CausalX into SingularTrajectory | SingularTrajectory · ETH/UCY ZARA2 | ADE: 0.149 -> 0.155 | Hurt |
| plug CausalX into TCNM | TCNM · Tropical cyclone 24h | distance error (km): 93.08 -> 74.55 | Improved |
| plug CausalX into TC-Diffuser | TC-Diffuser · Tropical cyclone 24h | distance error (km): 77.35 -> 73.76 | Improved |
| plug CausalX into TCNM for long horizon | TCNM · Tropical cyclone 72h | distance error (km): 408.35 -> 316.00 | Improved |
| add Granger causality constraint (ablation) | CausalX on TCNM · Tropical cyclone | distance error (km): 203.47 -> 181.74 | Improved |
| add do-calculus constraint (ablation) | CausalX on SingularTrajectory · ETH/UCY UNIV | FDE: 0.446 -> 0.429 | Improved |
| add TDMI constraint (ablation) | CausalX on TCNM · Tropical cyclone | distance error (km): 206.42 -> 181.74 | Improved |
| add VAE constraint (ablation) | CausalX on TCNM · Tropical cyclone | distance error (km): 200.78 -> 181.74 | Improved |
| add diffusion-based graph refinement (RefDif) (ablation) | CausalX on TCNM · Tropical cyclone | distance error (km): 189.43 -> 181.74 | Improved |
Closest papers by meaning
- Causal Spatio-Temporal Prediction: An Effective and Efficient Multi-Modal Approach (NeurIPS 2025 · similarity 0.93)
- Causal Deciphering and Inpainting in Spatio-Temporal Dynamics via Diffusion Model (NeurIPS 2024 · similarity 0.88)
- TimeXL: Explainable Multi-modal Time Series Prediction with LLM-in-the-Loop (NeurIPS 2025 · similarity 0.87)
- Discovering Latent Causal Graphs from Spatiotemporal Data (ICML 2025 · similarity 0.86)
- Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling (ICML 2024 · similarity 0.86)
- How Different from the Past? Spatio-Temporal Time Series Forecasting with Self-Supervised Deviation Learning (NeurIPS 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.
