发表机构
Jilin University; The Chinese University of Hong Kong, Shenzhen（吉林大学; 香港中文大学（深圳））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有表格生成器忽视语义约束的问题，提出语义一致表格扩散框架，利用LLM提取列内语义和列间规则作为生成条件，在统一语义空间进行条件去噪，在六个基准上同时提升分布保真度与语义一致性。
AI 中文摘要
合成表格数据可以匹配真实数据分布，但仍可能违反管理有效表格行的语义约束。这揭示了现有表格生成器的一个关键局限性：它们主要优化分布保真度，但并未显式建模表格模式和文本描述中编码的弱语义先验。在本文中，我们提出\ours，一个语义一致的表格扩散框架，用于在弱指定语义先验下生成高保真合成数据。\ours首先构建两类先验，即列内语义和列间符号规则，通过LLM辅助从元数据中提取并在真实训练划分上进行验证。这些先验随后被用作生成条件而非事后过滤器。具体而言，\ours将异构列值、列身份和语义先验映射到统一的语义空间，并执行逐列前向损坏和先验条件反向去噪，以同时保留边际分布和规则一致的跨列依赖。在六个真实世界表格基准上的广泛实验表明，\ours在分布保真度、语义一致性和下游任务效用方面持续优于代表性的基于VAE、GAN、LLM和扩散的基线。额外的分析进一步证明了\ours在语义先验部分不可用时的鲁棒性。
英文摘要
Synthetic tabular data can match real data distributions while still violating the semantic constraints that govern valid tabular rows. This reveals a key limitation of existing tabular generators: they mainly optimize distributional fidelity, but do not explicitly model weak semantic priors encoded in tabular schema and textual descriptions. In this paper, we propose \ours, a semantics-consistent tabular diffusion framework for high-fidelity synthetic data generation under weakly specified semantic priors. \ours\ first constructs two types of priors, namely intra-column semantics and inter-column symbolic rules, with LLM-assisted extraction from metadata and validation on the real training split. These priors are then used as generation conditions rather than post-hoc filters. Specifically, \ours\ maps heterogeneous column values, column identities, and semantic priors into a unified semantic space, and performs column-wise forward corruption and prior-conditioned reverse denoising to preserve both marginal distributions and rule-consistent cross-column dependencies. Extensive experiments on six real-world tabular benchmarks show that \ours\ consistently improves distributional fidelity, semantic consistency, and downstream task utility over representative VAE-, GAN-, LLM-, and diffusion-based baselines. Additional analyses further demonstrate the robustness of \ours\ when semantic priors are partially unavailable.
