Loading Now

Summary of Booksql: a Large Scale Text-to-sql Dataset For Accounting Domain, by Rahul Kumar and Amar Raja Dibbu and Shrutendra Harsola and Vignesh Subrahmaniam and Ashutosh Modi


BookSQL: A Large Scale Text-to-SQL Dataset for Accounting Domain

by Rahul Kumar, Amar Raja Dibbu, Shrutendra Harsola, Vignesh Subrahmaniam, Ashutosh Modi

First submitted to arxiv on: 12 Jun 2024

Categories

  • Main: Computation and Language (cs.CL)
  • Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

     Abstract of paper      PDF of paper


GrooveSquid.com Paper Summaries

GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!

Summary difficulty Written by Summary
High Paper authors High Difficulty Summary
Read the original abstract here
Medium GrooveSquid.com (original content) Medium Difficulty Summary
The proposed paper aims to address the lack of large-scale datasets for natural language interfaces to databases in essential domains like finance and accounting. The authors introduce a new dataset, BookSQL, comprising 100k natural language queries-SQL pairs and accounting databases with 1 million records. They experiment with state-of-the-art models, including GPT-4, on the BookSQL dataset, revealing significant performance gaps that suggest developing more focused models for this domain.
Low GrooveSquid.com (original content) Low Difficulty Summary
BookSQL is a new large-scale dataset designed to help people without technical backgrounds interact with accounting and financial databases using natural language queries. The dataset contains 100k pairs of questions and answers, along with an accounting database containing 1 million records. Researchers are trying different methods to see how well they work on this new dataset.

Keywords

» Artificial intelligence  » Gpt