Impute-EM：用于异构数据插补的原生混合状态扩散模型
Impute-EM: Native Mixed-State Diffusion Models for Heterogeneous Data Imputation
- Applied AI Institute（应用人工智能研究所）
- Moscow State University（莫斯科国立大学）
- Yandex Research（Yandex研究院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
Impute-EM提出期望最大化框架，交替插补缺失值与重拟合原生混合状态扩散模型，解决异构数据插补中离散与连续结构不匹配问题，实现最佳分布保真度。
AI中文摘要：
缺失值在异构数据挖掘中普遍存在，其中数值、类别和二元变量常常共存。许多插补方法，尤其是基于扩散的方法，通过连续替代物（如one-hot松弛）来处理离散变量，而非原生地对它们建模。这造成了模型状态空间与数据的混合离散和连续结构之间的不匹配。我们提出Impute-EM，一个期望最大化框架，它在使用当前模型插补缺失条目和基于完整数据重新拟合扩散骨干之间交替进行。我们用用于异构数据的原生混合状态扩散骨干实例化Impute-EM，结合高斯和掩码类别组件，无需one-hot松弛。在精确设置中，我们刻画了更新过程，并表明在极限情况下，观测到的掩码索引边际与目标匹配，同时明确指出，仅凭不完整观测通常无法识别完整数据分布。实验上，Impute-EM在混合类型表格插补中提供了最佳的分布保真度，下游建模依赖于此，而文本插补作为原生离散骨干的受控验证。
英文摘要：
Missing values are ubiquitous in heterogeneous data mining, where numerical, categorical, and binary variables often coexist. Many imputation methods, especially diffusion-based ones, treat discrete variables through continuous surrogates such as one-hot relaxations rather than modeling them natively. This creates a mismatch between the model state space and the mixed discrete and continuous structure of the data. We propose Impute-EM, an Expectation Maximization style framework that alternates between imputing missing entries with the current model and refitting a diffusion backbone on completed data. We instantiate Impute-EM with native mixed-state diffusion backbones for heterogeneous data, combining Gaussian and masked categorical components without one-hot relaxations. In exact settings, we characterize the update and show that the observed mask-indexed marginals match the targets at the limit, while making explicit that the full data distribution is generally non-identifiable from incomplete observations alone. Empirically, Impute-EM delivers the best distributional fidelity on mixed-type tabular imputation, on which downstream modeling relies, with text imputation serving as a controlled validation of the native discrete backbone.
