HyperJoin: LLM-Enhanced Hypergraph Approach to Joinable Table Discovery
Published:Jan 6, 2026 05:00
•1 min read
•ArXiv NLP
Analysis
This paper introduces a novel approach to joinable table discovery by leveraging LLMs and hypergraphs to capture complex relationships between tables and columns. The proposed HyperJoin framework addresses limitations of existing methods by incorporating both intra-table and inter-table structural information, potentially leading to more coherent and accurate join results. The use of a hierarchical interaction network and coherence-aware reranking module are key innovations.
Key Takeaways
- •HyperJoin uses a hypergraph to model tables and their relationships.
- •It employs a Hierarchical Interaction Network (HIN) for column representation learning.
- •A coherence-aware reranking module improves the consistency of join results.
Reference
“To address these limitations, we propose HyperJoin, a large language model (LLM)-augmented Hypergraph framework for Joinable table discovery.”