Researchers have proposed a new method for imputing missing values in time series data. The Masked Diffusion Time-series Imputation Model (MDTIM) uses a masked diffusion model to predict the original signal rather than added noise. To handle continuous data, the authors introduce Stochastic Discretization, which maps continuous values to ordinal-aware tokens while preserving dynamics. Experiments on various benchmarks show that MDTIM outperforms state-of-the-art baselines in terms of robustness and scalability. --- Why it matters: This matters for engineers working with time series data because it provides a more accurate and efficient way to handle missing values, which is crucial for reliable analysis. The proposed method can be applied to various domains where time series data is used, such as finance, healthcare, or energy management. Source: https://arxiv.org/abs/2608.19119