Research map / Time series / Causal and generative time series
DoFlow: Flow-based Generative Models for Interventional and Counterfactual Forecasting on Time Series
Dongze Wu, Feng Qiu, Yao Xie
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 |
|---|---|---|---|
| use DoFlow instead of adapted forecasters for interventional queries | DoFlow · Tree (additive) | RMSE (interventional): 0.74 -> 0.54 | Improved |
| use DoFlow instead of adapted forecasters for interventional queries | DoFlow · FC-Layer (NLNA) | RMSE (interventional): 1.02 -> 0.36 | Improved |
| use DoFlow for observational forecasting vs strongest baseline | DoFlow · Diamond (additive) | RMSE (observational): 0.49 -> 0.55 | Hurt |
| use DoFlow for observational forecasting vs strongest baseline | DoFlow · FC-Layer (additive) | RMSE (observational): 0.38 -> 0.39 | No clear effect |
| use DoFlow for observational forecasting vs strongest baseline | DoFlow · Diamond (NLNA) | RMSE (observational): 0.33 -> 0.31 | Improved |
| use DoFlow encode-decode counterfactual generation under NLNA SCMs outside the theory assumptions | DoFlow · Tree (NLNA) | RMSE (counterfactual): 0.11 -> 0.14 | No clear effect |
Closest papers by meaning
- Learning Structural Causal Models from Ordering: Identifiable Flow Models (AAAI 2025 · similarity 0.86)
- TimeXL: Explainable Multi-modal Time Series Prediction with LLM-in-the-Loop (NeurIPS 2025 · similarity 0.85)
- TARFVAE: Efficient One-Step Generative Time Series Forecasting via TARFLOW based VAE (NeurIPS 2025 · similarity 0.85)
- Causally Consistent Normalizing Flow (AAAI 2025 · similarity 0.85)
- Universal Time Series Generation with Neural Controlled Differential Equations (NeurIPS 2026 · similarity 0.85)
- Graph Neural Flows for Unveiling Systemic Interactions Among Irregularly Sampled Time Series (NeurIPS 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.
