Research map / Probabilistic and scientific modeling / Dynamical systems
Decomposing Temporal High-Order Interactions via Latent ODEs
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use latent ODE decomposition (THIS-ODE) instead of discrete-step or GP/NN time-aware baselines | temporal high-order interaction decomposition · Real-world datasets, interpolation and extrapolation (r=3), RMSE | RMSE | Helped |
| Time alignment (integral transform to common end time) in mini-batch ODE solving (ablation) | THIS-ODE state computation · Spiral simulation, RK5 solver | relative state error: 1.1 vs 1.2 ×10^-4 | No clear effect |
| Time alignment in mini-batch ODE solving (ablation) | THIS-ODE sensitivity computation · Spiral simulation, RK5 solver | relative sensitivity error: 4.2 vs 4.1 ×10^-5 | No clear effect |
| Discrete time steps with GP tensor factorization (GPTF-DTN) and similar discrete models as baselines | temporal decomposition baselines · Server Room interpolation, r=3 | RMSE: 0.373 vs 0.246 for THIS-ODE | Hurt |
| Gaussian-process time-as-input GPTF-Time and NTF-Time vs THIS-ODE in spiral simulation | spiral simulation · Spiral interpolation/extrapolation | RMSE: GPTF-Time 0.5557 / NTF-Time 0.1004 vs THIS-ODE 0.0148 (interpolation) | Hurt |
| Using THIS-ODE with r=2 instead of GPTF-Time on Beijing Air interpolation | THIS-ODE · Beijing Air interpolation, r=2 | RMSE: 0.521 (GPTF-Time) vs 0.524 (THIS-ODE) | Mixed |
Closest papers by meaning
- HOPE: High-order Graph ODE For Modeling Interacting Dynamics (ICML 2023 · similarity 0.90)
- Functional Complexity-adaptive Temporal Tensor Decomposition (NeurIPS 2025 · similarity 0.86)
- Physics-informed Reduced Order Modeling of Time-dependent PDEs via Differentiable Solvers (NeurIPS 2025 · similarity 0.85)
- PGODE: Towards High-quality System Dynamics Modeling (ICML 2024 · similarity 0.85)
- CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs (NeurIPS 2025 · similarity 0.84)
- Graph Mamba Operator: A Latent Simulator for Interacting Particle Systems (NeurIPS 2026 · similarity 0.84)
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.