arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.01015cs.CLcs.DB
收藏

HyperJoin: 基于大语言模型的超图链接预测用于可连接表发现

HyperJoin: LLM-augmented Hypergraph Link Prediction for Joinable Table Discovery

University of Technology Sydney(悉尼科技大学);University of New South Wales(新南威尔士大学);ACEM, Shanghai Jiao Tong University(ACEM,上海交通大学)

Shiyuan Liu, Jianwei Wang, Xuemin Lin, Lu Qin, Wenjie Zhang, Ying Zhang

AI总结:

HyperJoin通过构建超图并利用大语言模型增强的超图链接预测,提升可连接表发现的精度和召回率。

摘要:

作为数据湖管理中的关键任务,可连接表发现已引起广泛关注。尽管现有基于语言模型的方法通过结合离线列表示学习与在线排序取得了显著成效,但其设计在底层结构交互方面考虑不足:(1)离线阶段,它们直接将表建模为孤立或成对的列,从而难以捕捉丰富的跨表和内表结构信息;(2)在线阶段,它们仅基于查询-候选相似性对候选列进行排序,忽略了候选者之间的相互作用,导致结果集不一致。为解决这些局限性,我们提出了HyperJoin,一种用于可连接表发现的基于大语言模型(LLM)的超图框架。具体而言,我们首先构建超图以建模表,利用内表超边和LLM增强的跨表超边。因此,可连接表发现的任务被建模为在此构建的超图上的链接预测。随后,我们设计了HIN,一种分层交互网络,通过列和超边之间的双向信息传递学习表达性列表示。为了加强结果列的连贯性和内部一致性,我们将在线排序建模为一个具有连贯性的top-k列选择问题。然后,我们引入了一个重新排序模块,利用最大生成树算法来修剪噪声连接并最大化连贯性。实验表明,HyperJoin的优越性,相对于最佳基线,平均提高了21.4%(Precision@15)和17.2%(Recall@15).

英文摘要:

As a pivotal task in data lake management, joinable table discovery has attracted widespread interest. While existing language model-based methods achieve remarkable performance by combining offline column representation learning with online ranking, their design insufficiently accounts for the underlying structural interactions: (1) offline, they directly model tables into isolated or pairwise columns, thereby struggling to capture the rich inter-table and intra-table structural information; and (2) online, they rank candidate columns based solely on query-candidate similarity, ignoring the mutual interactions among the candidates, leading to incoherent result sets. To address these limitations, we propose HyperJoin, a large language model (LLM)-augmented Hypergraph framework for Joinable table discovery. Specifically, we first construct a hypergraph to model tables using both the intra-table hyperedges and the LLM-augmented inter-table hyperedges. Consequently, the task of joinable table discovery is formulated as link prediction on this constructed hypergraph. We then design HIN, a Hierarchical Interaction Network that learns expressive column representations through bidirectional message passing over columns and hyperedges. To strengthen coherence and internal consistency in the result columns, we cast online ranking as a coherence-aware top-k column selection problem. We then introduce a reranking module that leverages a maximum spanning tree algorithm to prune noisy connections and maximize coherence. Experiments demonstrate the superiority of HyperJoin, achieving average improvements of 21.4% (Precision@15) and 17.2% (Recall@15) over the best baseline.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑