.png){kind=icon}
.png){kind=icon}
Return
Diffusion Denoised and Physics-regularized Inter-series Model for Long-horizon Multivariate Time-series Forecasting
.png){kind=icon}
OA
AI
DOI：10.1016/j.inffus.2026.104325.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Score-based denoising raises SNR for reliable cross-series statistics. • Dynamic, thresholded graphs capture regime-dependent sparse couplings. • RD prior stabilizes multi-step rollouts via soft physics-inspired limits. • Theory: contractive horizon map and Lipschitz-controlled graph blocks. • SOTA accuracy on six LTSF benchmarks under standardized budgets.
Keywords:
Time-series forecasting
Diffusion Denoising
Dynamic Correlation Graphs
Physics Regularization
Reaction-diffusion Stabilization
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
15.5
Papers:
4.1K
Citations:
2.7W