time-series-imputation

The paper proposes clipping-aware objective conditioning for differentially private diffusion models applied to energy time-series imputation, using v-prediction and timestep-aware loss weighting to reduce pre-clipping gradient amplification and improve imputation utility over the epsilon-prediction baseline on five real-world energy datasets.

GAUDI proposes a conditional diffusion model that leverages missingness geometry for imputing contiguous missing blocks in air-quality time-series data, showing improved performance over existing methods.

This paper introduces CAIR, a two-stage framework for imputing physiological time-series data under realistic missingness, outperforming existing methods by incorporating gap mechanisms and curriculum-aware training.