FINCH: Financial Intelligence using Natural language for Contextualized SQL Handling

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Singh, Avinash Kumar, Sarmah, Bhaskarjit, Pasquali, Stefano
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916985517899776
author Singh, Avinash Kumar
Sarmah, Bhaskarjit
Pasquali, Stefano
author_facet Singh, Avinash Kumar
Sarmah, Bhaskarjit
Pasquali, Stefano
contents Text-to-SQL, the task of translating natural language questions into SQL queries, has long been a central challenge in NLP. While progress has been significant, applying it to the financial domain remains especially difficult due to complex schema, domain-specific terminology, and high stakes of error. Despite this, there is no dedicated large-scale financial dataset to advance research, creating a critical gap. To address this, we introduce a curated financial dataset (FINCH) comprising 292 tables and 75,725 natural language-SQL pairs, enabling both fine-tuning and rigorous evaluation. Building on this resource, we benchmark reasoning models and language models of varying scales, providing a systematic analysis of their strengths and limitations in financial Text-to-SQL tasks. Finally, we propose a finance-oriented evaluation metric (FINCH Score) that captures nuances overlooked by existing measures, offering a more faithful assessment of model performance.
format Preprint
id arxiv_https___arxiv_org_abs_2510_01887
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FINCH: Financial Intelligence using Natural language for Contextualized SQL Handling
Singh, Avinash Kumar
Sarmah, Bhaskarjit
Pasquali, Stefano
Computational Finance
Artificial Intelligence
Text-to-SQL, the task of translating natural language questions into SQL queries, has long been a central challenge in NLP. While progress has been significant, applying it to the financial domain remains especially difficult due to complex schema, domain-specific terminology, and high stakes of error. Despite this, there is no dedicated large-scale financial dataset to advance research, creating a critical gap. To address this, we introduce a curated financial dataset (FINCH) comprising 292 tables and 75,725 natural language-SQL pairs, enabling both fine-tuning and rigorous evaluation. Building on this resource, we benchmark reasoning models and language models of varying scales, providing a systematic analysis of their strengths and limitations in financial Text-to-SQL tasks. Finally, we propose a finance-oriented evaluation metric (FINCH Score) that captures nuances overlooked by existing measures, offering a more faithful assessment of model performance.
title FINCH: Financial Intelligence using Natural language for Contextualized SQL Handling
topic Computational Finance
Artificial Intelligence
url https://arxiv.org/abs/2510.01887