We introduce SMUTF, a unique approach for large-scale tabular data schema matching (SM), which assumes that supervised learning does not affect performance in open-domain tasks, thereby enabling effective cross-domain matching. This system uniquely combines rule-based feature engineering, pre-trained language models, and generative large language models. In an innovative adaptation inspired by the Humanitarian Exchange Language, we deploy 'generative tags' for each data column, enhancing the effectiveness of SM. SMUTF exhibits extensive versatility, working seamlessly with any pre-existing pre-trained embeddings, classification methods, and generative models. Recognizing the lack of extensive, publicly available datasets for SM, we have created and open-sourced the HDXSM dataset from the public humanitarian data. We believe this to be the most exhaustive SM dataset currently available. In evaluations across various public datasets and the novel HDXSM dataset, SMUTF demonstrated exceptional performance, surpassing existing state-of-the-art models in terms of accuracy and efficiency, and} improving the F1 score by 11.84% and the AUC of ROC by 5.08%.

翻译：本文提出SMUTF方法，一种用于大规模表格数据的模式匹配（SM）创新方案，其核心假设是监督学习不影响开放域任务的性能，从而实现有效的跨域匹配。该体系独特地融合了基于规则的特征工程、预训练语言模型与生成式大语言模型。受人道主义交换语言（HXL）创新适配方法的启发，我们为每个数据列部署"生成标签"，显著提升模式匹配效果。SMUTF展现出广泛通用性，可无缝适配任意现有预训练嵌入、分类方法及生成模型。针对当前缺乏大规模公开模式匹配数据集的现状，我们从公开人道主义数据中构建并开源了HDXSM数据集，我们相信这是目前最全面的模式匹配数据集。在多个公开数据集及HDXSM新数据集上的评估显示，SMUTF在准确率和效率上均超越现有最优模型，F1分数提升11.84%，ROC曲线下面积（AUC）提升5.08%。