Research map / Time series / Causal and generative time series
Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglement
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 D3VAE instead of the strongest listed baseline | Time-series forecasting · Traffic, input-8-predict-8 | MSE: 0.794 -> 0.081 | Helped |
| Use D3VAE instead of the strongest listed baseline | Time-series forecasting · Electricity, input-8-predict-8 | MSE: 0.853 -> 0.251 | Helped |
| Use D3VAE instead of the strongest listed baseline | Time-series forecasting · Weather, input-8-predict-8 | MSE: 0.234 -> 0.169 | Helped |
| Use D3VAE instead of the strongest listed baseline | Time-series forecasting · ETTm1, input-8-predict-8 | MSE: 0.921 -> 0.527 | Helped |
| Use D3VAE instead of the strongest listed baseline | Time-series forecasting · ETTh1, input-8-predict-8 | CRPS: 0.461 -> 0.424 | Helped |
| Use D3VAE instead of the strongest listed baseline | Time-series forecasting · Wind, input-8-predict-8 | MSE: 1.321 -> 0.681 | Helped |
| Add diffusion of target Y (ablation) | D3VAE · Traffic, input-16-predict-16 | MSE: 0.122 -> 0.081 | Helped |
| Add diffusion of target Y (ablation) | D3VAE · Traffic, input-16-predict-16 | CRPS: 0.250 -> 0.200 | Helped |
| Add diffusion of input X (ablation) | D3VAE · Traffic, input-16-predict-16 | MSE: 0.123 -> 0.081 | Helped |
| Add coupled diffusion (ablation) | D3VAE · Traffic, input-16-predict-16 | CRPS: 0.255 -> 0.200 | Helped |
| Add DSM when the target is not diffused (ablation) | D3VAE without diffused target · Traffic, input-16-predict-16 | MSE: 0.096 -> 0.122 | Mixed |
| Use a single-step denoising jump rather than more sampling steps | D3VAE · Time-series forecasting | prediction performance | No clear effect |
Closest papers by meaning
- Diffusion-TS: Interpretable Diffusion for General Time Series Generation (ICLR 2024 · similarity 0.91)
- Latent Diffusion Transformer for Probabilistic Time Series Forecasting (AAAI 2024 · similarity 0.90)
- Probabilistic Time Series Modeling with Decomposable Denoising Diffusion Model (ICML 2024 · similarity 0.90)
- TimeDART: A Diffusion Autoregressive Transformer for Self-Supervised Time Series Representation (ICML 2025 · similarity 0.89)
- Conditional Diffusion Model with Nonlinear Data Transformation for Time Series Forecasting (ICML 2025 · similarity 0.89)
- TARFVAE: Efficient One-Step Generative Time Series Forecasting via TARFLOW based VAE (NeurIPS 2025 · similarity 0.89)
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.