SQUiD: Synthesizing Relational Databases from Unstructured Text

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sadia, Mushtari, Yang, Zhenning, Xiao, Yunming, Chen, Ang, Chowdhury, Amrita Roy
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917300973600768
author Sadia, Mushtari
Yang, Zhenning
Xiao, Yunming
Chen, Ang
Chowdhury, Amrita Roy
author_facet Sadia, Mushtari
Yang, Zhenning
Xiao, Yunming
Chen, Ang
Chowdhury, Amrita Roy
contents Relational databases are central to modern data management, yet most data exists in unstructured forms like text documents. To bridge this gap, we leverage large language models (LLMs) to automatically synthesize a relational database by generating its schema and populating its tables from raw text. We introduce SQUiD, a novel neurosymbolic framework that decomposes this task into four stages, each with specialized techniques. Our experiments show that SQUiD consistently outperforms baselines across diverse datasets. Our code and datasets are publicly available at: https://github.com/Mushtari-Sadia/SQUiD.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19025
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SQUiD: Synthesizing Relational Databases from Unstructured Text
Sadia, Mushtari
Yang, Zhenning
Xiao, Yunming
Chen, Ang
Chowdhury, Amrita Roy
Databases
Computation and Language
Relational databases are central to modern data management, yet most data exists in unstructured forms like text documents. To bridge this gap, we leverage large language models (LLMs) to automatically synthesize a relational database by generating its schema and populating its tables from raw text. We introduce SQUiD, a novel neurosymbolic framework that decomposes this task into four stages, each with specialized techniques. Our experiments show that SQUiD consistently outperforms baselines across diverse datasets. Our code and datasets are publicly available at: https://github.com/Mushtari-Sadia/SQUiD.
title SQUiD: Synthesizing Relational Databases from Unstructured Text
topic Databases
Computation and Language
url https://arxiv.org/abs/2505.19025