AI 中文总结
FUSE是一种用于混合类型表格流匹配的方法，通过分离特征处理与跨列交互，在8个表格数据集上实现了分布保真度和下游效用的强劲一致性能。
AI 中文摘要
生成混合类型表格数据需要联合建模不同特征分布及其复杂的跨列依赖关系。变分流匹配通过分解分布处理不同端点，但将特征特定处理和跨列交互隐含在共享主干中。我们提出特征级统一专业化与跨列交换（FUSE）以明确分离这些角色。FUSE对数值特征和分类特征应用独立的自适应混合模块，使每个特征可结合共享的专业化子网络，同时联合注意力保留所有列间的信息交换。我们还刻画了受限条件上下文导致的总体风险过剩，并将连续Wasserstein生成误差以端点预测风险为界。在8个表格数据集上的综合实验表明，FUSE在分布保真度和下游效用指标上均实现了强劲且一致的性能。
英文摘要
Generating mixed-type tabular data requires jointly modeling diverse feature distributions and their complex cross-column dependencies. Variational flow matching handles distinct endpoints via factorized distributions, yet leaves feature-specific processing and cross-column interactions implicit within a shared backbone. We introduce Feature-wise Unified Specialization with cross-column Exchange (FUSE) to explicitly separate these roles. FUSE applies separate adaptive mixture modules to numerical and categorical features, allowing each feature to combine shared specialized subnetworks, while joint attention preserves information exchange across all columns. We also characterize the excess population risk from restricted conditioning contexts and bound the continuous Wasserstein generation error by endpoint-prediction risk. Comprehensive experiments on eight tabular datasets demonstrate that FUSE achieves strong and consistent performance across distributional fidelity and downstream utility metrics.
Comments19 pages, 7 figures, 7 tables
