DSDI combines conditional diffusion, a weight-reducing AR prediction injection, and a multi-scale S4 U-Net to improve multivariate time series imputation, outperforming baselines on DACMI, ETT, and AQI.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Boundary-enhanced time series data imputation with long-term dependency diffusion models
DSDI combines conditional diffusion, a weight-reducing AR prediction injection, and a multi-scale S4 U-Net to improve multivariate time series imputation, outperforming baselines on DACMI, ETT, and AQI.