Text2VectorSQL: Towards a Unified Interface for Vector Search and SQL Queries

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wang, Zhengren, Yao, Dongwen, Li, Bozhou, Ma, Dongsheng, Li, Bo, Li, Zhiyu, Xiong, Feiyu, Cui, Bin, Tang, Linpeng, Zhang, Wentao
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866915602517458944
author	Wang, Zhengren Yao, Dongwen Li, Bozhou Ma, Dongsheng Li, Bo Li, Zhiyu Xiong, Feiyu Cui, Bin Tang, Linpeng Zhang, Wentao
author_facet	Wang, Zhengren Yao, Dongwen Li, Bozhou Ma, Dongsheng Li, Bo Li, Zhiyu Xiong, Feiyu Cui, Bin Tang, Linpeng Zhang, Wentao
contents	The proliferation of unstructured data poses a fundamental challenge to traditional database interfaces. While Text-to-SQL has democratized access to structured data, it remains incapable of interpreting semantic or multi-modal queries. Concurrently, vector search has emerged as the de facto standard for querying unstructured data, but its integration with SQL-termed VectorSQL-still relies on manual query crafting and lacks standardized evaluation methodologies, creating a significant gap between its potential and practical application. To bridge this fundamental gap, we introduce and formalize Text2VectorSQL, a novel task to establish a unified natural language interface for seamlessly querying both structured and unstructured data. To catalyze research in this new domain, we present a comprehensive foundational ecosystem, including: (1) A scalable and robust pipeline for synthesizing high-quality Text-to-VectorSQL training data. (2) VectorSQLBench, the first large-scale, multi-faceted benchmark for this task, encompassing 12 distinct combinations across three database backends (SQLite, PostgreSQL, ClickHouse) and four data sources (BIRD, Spider, arXiv, Wikipedia). (3) Several novel evaluation metrics designed for more nuanced performance analysis. Extensive experiments not only confirm strong baseline performance with our trained models, but also reveal the recall degradation challenge: the integration of SQL filters with vector search can lead to more pronounced result omissions than in conventional filtered vector search. By defining the core task, delivering the essential data and evaluation infrastructure, and identifying key research challenges, our work lays the essential groundwork to build the next generation of unified and intelligent data interfaces. Our repository is available at https://github.com/OpenDCAI/Text2VectorSQL.
format	Preprint
id	arxiv_https___arxiv_org_abs_2506_23071
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	Text2VectorSQL: Towards a Unified Interface for Vector Search and SQL Queries Wang, Zhengren Yao, Dongwen Li, Bozhou Ma, Dongsheng Li, Bo Li, Zhiyu Xiong, Feiyu Cui, Bin Tang, Linpeng Zhang, Wentao Computation and Language The proliferation of unstructured data poses a fundamental challenge to traditional database interfaces. While Text-to-SQL has democratized access to structured data, it remains incapable of interpreting semantic or multi-modal queries. Concurrently, vector search has emerged as the de facto standard for querying unstructured data, but its integration with SQL-termed VectorSQL-still relies on manual query crafting and lacks standardized evaluation methodologies, creating a significant gap between its potential and practical application. To bridge this fundamental gap, we introduce and formalize Text2VectorSQL, a novel task to establish a unified natural language interface for seamlessly querying both structured and unstructured data. To catalyze research in this new domain, we present a comprehensive foundational ecosystem, including: (1) A scalable and robust pipeline for synthesizing high-quality Text-to-VectorSQL training data. (2) VectorSQLBench, the first large-scale, multi-faceted benchmark for this task, encompassing 12 distinct combinations across three database backends (SQLite, PostgreSQL, ClickHouse) and four data sources (BIRD, Spider, arXiv, Wikipedia). (3) Several novel evaluation metrics designed for more nuanced performance analysis. Extensive experiments not only confirm strong baseline performance with our trained models, but also reveal the recall degradation challenge: the integration of SQL filters with vector search can lead to more pronounced result omissions than in conventional filtered vector search. By defining the core task, delivering the essential data and evaluation infrastructure, and identifying key research challenges, our work lays the essential groundwork to build the next generation of unified and intelligent data interfaces. Our repository is available at https://github.com/OpenDCAI/Text2VectorSQL.
title	Text2VectorSQL: Towards a Unified Interface for Vector Search and SQL Queries
topic	Computation and Language
url	https://arxiv.org/abs/2506.23071