Research map / Probabilistic and scientific modeling / Molecules and proteins
Spatiotemporal forecasting and imputation
Forecasts or imputes spatiotemporal data with graph-informed flow matching and residual diffusion.
30 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Molecules and proteins, keyword cluster “flow · trajectory · matching”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
6 papers in 2023–24 and 14 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 1.2% to 1.5%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Fortune Teller: Predicting Your Career Path (AAAI 2016 · 257 citations)
- 2018 · Automatically Inferring Data Quality for Spatiotemporal Forecasting (ICLR 2018 · 4 citations)
- 2019 · Regression Planning Networks (NeurIPS 2019 · 9 citations)
- 2020 · Learning to Select Best Forecast Tasks for Clinical Outcome Prediction (NeurIPS 2020 · 5 citations)
- 2021 · Modeling Heterogeneous Relations across Multiple Modes for Potential Crowd Flow Prediction (AAAI 2021 · 23 citations)
- 2022 · STDEN: Towards Physics-Guided Neural Networks for Traffic Flow Prediction (AAAI 2022 · 106 citations)
- 2024 · DiffCast: A Unified Framework via Residual Diffusion for Precipitation Nowcasting (CVPR 2024 · 68 citations)
- 2025 · VQLTI: Long-Term Tropical Cyclone Intensity Forecasting with Physical Constraints (AAAI 2025 · 7 citations)
- 2026 · Detecting the Future: All-at-Once Event Sequence Forecasting with Horizon Matching (AAAI 2026 · 1 citation)
Earliest papers here
- Fortune Teller: Predicting Your Career Path (AAAI 2016 · 257 citations)
- Disease Trajectory Maps (NeurIPS 2016 · 2 citations)
- Automatically Inferring Data Quality for Spatiotemporal Forecasting (ICLR 2018 · 4 citations)
What has been tried
24 papers here report 246 measured changes (a component added, removed or swapped, compared with a baseline): helped 70%, no clear effect 6%, hurt 20%, mixed 5%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Set α=1, removing the deterministic loss | SEVIR · CSI | 0.3077 -> 0.2430 | Hurt | DiffCast: A Unified Framework via Residual Diffusion for Precipitation Nowcasting (CVPR 2024 · 68 citations) |
| add the ASA module rather than use a sum operator | Flight trajectory test set, horizon 15 · MAE (longitude) | 0.0123 -> 0.0124 | No clear effect | FlightBERT++: A Non-autoregressive Multi-Horizon Flight Trajectory Prediction Framework (AAAI 2024 · 32 citations) |
| Add codebook quantization | Global 2022, 120h · MSLP MAE | 25.10 -> 21.45 | Helped | VQLTI: Long-Term Tropical Cyclone Intensity Forecasting with Physical Constraints (AAAI 2025 · 7 citations) |
| Add vehicle length, width, and height as flow features | DrivAerNet++ · MSE (×10−5) | 5.293 -> 5.275 | Helped | Aerodynamic Coefficients Prediction via Cross-Attention Fusion and Physical-Informed Training (AAAI 2025 · 6 citations) |
| Replace AMSE with MSE | Fine-scale forecast variability · spectral sharpness | Hurt | Fixing the Double Penalty in Data-Driven Weather Forecasting Through a Modified Spherical Harmonic Loss Function (ICML 2025 · 5 citations) | |
| add sequential trend (ATU-based) module | Temperature and relative humidity · MSE | temperature 1.46 -> 0.87; humidity 45.82 -> 36.97 | Helped | iTrendRNN: An Interpretable Trend-Aware RNN for Meteorological Spatiotemporal Prediction (AAAI 2024 · 5 citations) |
| Add text inputs | Datasets with text modalities · prediction accuracy | Helped | Causal Spatio-Temporal Prediction: An Effective and Efficient Multi-Modal Approach (NeurIPS 2025 · 3 citations) | |
| Add approaching-infeasible cases to training | Close-to-infeasible cases, Task 4.2 · power balance loss | Helped | PF∆: A Benchmark Dataset for Power Flow under Load, Generation, and Topology Variations (NeurIPS 2025 · 1 citation) | |
| Add covariates to Linear | Broll ID · RMSE | 11.68 -> 9.95 | Helped | GlucoBench: Curated List of Continuous Glucose Monitoring Datasets with Prediction Benchmarks (ICLR 2024 · 1 citation) |
| Add AWF to the baseline | Anesthetized Dataset · Abs Rel | 0.393 -> 0.344 | Helped | Blood Flow Speed Estimation with Optical Coherence Tomography Angiography Images (CVPR 2025 · 0 citations) |
| Use a straight conditional path in TD-CFM(c) | 19 tasks, 4 domains · value-function MSE | Helped | Temporal Difference Flows (ICML 2025 · 0 citations) | |
| Add Hybrid-Balance to AGFN | CVRP200, CVRP500, CVRP1000 · Gap(%) | 11.48 -> 10.84; 12.21 -> 10.53; 11.15 -> 9.47 | Helped | Hybrid-Balance GFlowNet for Solving Vehicle Routing Problems (NeurIPS 2025 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 9 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- IBTrACS · 2 of 9 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 20 papers; from OpenAlex and the papers' affiliations).
- Harbin Institute of Technology · 3 papers
Use it
Download BibTeX of 29 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Flow matching generative modeling (97 papers)
- Multimodal trajectory prediction (95 papers)
Elsewhere in the topic
- Generative protein structure design (142 papers)
- Prediction-powered and conformal inference (129 papers)
- Statistical analysis of ML practice (122 papers)
- Neuro-symbolic probabilistic layers (120 papers)
- Structure-based 3D molecule generation (107 papers)
- Sparse graphical and tensor models (97 papers)
- Machine learning for physical simulation (96 papers)
- Probabilistic embeddings and relational inference (93 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).