[论文解读] Efficient and Effective Table-Centric Table Union Search in Data Lakes

Yongkang Sun, Zhihao Ding+3 more|arXiv (Cornell University)|Mar 18, 2026

Data Quality and Management被引用 0

一句话总结

TACTUS 引入一种以表为中心的顶部-k 表联合检索方法，先学习表嵌入以捕捉整体表级联合性，再用列级线索进行最终排序，在准确性和速度上优于以列为中心的方法。

ABSTRACT

In data lakes, information on the same subject is often fragmented across multiple tables. Table union search aims to find the top-k tables that can be unioned with a query table to extend it with more rows, without relying on metadata or ground-truth labels. Existing methods are mainly column-centric: they focus on modeling column unionability scores using column embeddings, which are then used throughout the search process for column matching, filtering, and aggregation. However, this overlooks holistic table-level semantics, which may result in suboptimal rankings and inefficiencies. We introduce TACTUS, a novel table-centric method for table union search. Unlike prior work that searches from columns to tables, we search in a table-first way and examine columns only in the final step. During offline processing, we directly generate table embeddings for holistic, table-level unionability scoring by designing table-level representation techniques, including positive table pair construction to simulate unionable tables, two-pronged negative table sampling to avoid latent positives and mine hard negatives to enhance representation quality, and attentive table encoding for effective embeddings. During online search, we first develop a table-centric adaptive candidate retrieval method that efficiently selects a compact, high-quality candidate pool by leveraging the distribution of table-level unionability scores induced by table embeddings. We then inspect columns only within this compact candidate set and design a dual-evidence reranking technique that integrates table-level and column-level scores to refine the final top-k results. Extensive experiments on real-world datasets show that TACTUS significantly improves result quality while being much faster than existing methods in both offline and online processing, often by an order of magnitude.

研究动机与目标

• 在数据湖中无需元数据即可搜索可联合的表的动机与必要性。
• 提出一个表为中心的框架，以捕捉用于联合性的整体表级语义。
• 开发离线表嵌入学习，利用自监督信号区分可联合与不可联合的表。
• 设计一个在线检索与重排序流程，将表级证据与列级证据结合以获得前-k结果。

提出的方法

• 为数据湖中的每个表 Ti 产生一个表嵌入 Ti，以编码表级联合性语义。
• 离线：构建正样本表对和双重负采样，通过表级对比损失训练一个具注意力的表编码器。
• 在线：使用表嵌入进行表中心的自适应候选检索，形成紧凑的候选集 S；再用结合 mu_T 与 mu_A 的双证据分数进行重排序。
• Mu_T 是来自嵌入的表级联合性分数；mu_A 是离线通过基于 FastText 的特征获得的列对齐分数，合成为 mu 以进行最终排序。）
•  

实验结果

研究问题

• RQ1 如何在不依赖表元数据的情况下，用表中心表示捕捉整体表级联合性？
• RQ2 两阶段流程（表级检索再进行列级精 refin）是否在准确性和效率上优于列中心方法？
• RQ3 哪些自监督策略最能近似可联合表对来训练表嵌入？
• RQ4 自适应候选检索是否能产生一个紧凑且覆盖率高的集合，从而在不牺牲前-k 质量的前提下加速在线搜索？

主要发现

• TACTUS 在所报告的实验中持续实现优越的效果性，相较于现有方法表现更好。
• 表中心流水线在离线和在线处理上均显著提速，且通常达到数量级的提升。
• 每个表一个表嵌入即可实现高效的候选检索并提前对非相关表进行裁剪。
• 双证据重排序有效地将表级嵌入与列级对齐结合起来，以优化前-k 结果。
• 离线的自监督表嵌入通过正样本对构建和困难负样本采样提升了联合性估计的质量。