# Synthetic SQL Column Descriptions and Their Impact on Text-to-SQL Performance

arXiv: https://arxiv.org/abs/2408.04691 · HTML: https://mlsanity.com/paper/2408.04691/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 24 references that are in the map.
- First posted on arXiv 8 Aug 2024.

## Builds on (most cited first, 18 of 18)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task](https://mlsanity.com/paper/1809.08887/) (2018)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning](https://mlsanity.com/paper/1709.00103/) (2017)
- [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://mlsanity.com/paper/2004.02349/) (2020)
- [UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models](https://mlsanity.com/paper/2201.05966/) (2022)
- [TableFormer: Robust Transformer Modeling for Table-Text Encoding](https://mlsanity.com/paper/2203.00274/) (2022)
- [Before Generation, Align it! A Novel and Effective Strategy for Mitigating Hallucinations in Text-to-SQL Generation](https://mlsanity.com/paper/2405.15307/) (2024)
- [Knowledge-to-SQL: Enhancing SQL Generation with Data Expert LLM](https://mlsanity.com/paper/2402.11517/) (2024)
- [MultiTabQA: Generating Tabular Answers for Multi-Table Question Answering](https://mlsanity.com/paper/2305.12820/) (2023)
- [Natural Language Interfaces for Tabular Data Querying and Visualization: A Survey](https://mlsanity.com/paper/2310.17894/) (2023)
- [TCN: Table Convolutional Network for Web Table Interpretation](https://mlsanity.com/paper/2102.09460/) (2021)
- [Table-to-Text Generation with Effective Hierarchical Encoder on Three Dimensions (Row, Column and Time)](https://mlsanity.com/paper/1909.02304/) (2019)
- [Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge](https://mlsanity.com/paper/2301.01067/) (2023)
- [NameGuess: Column Name Expansion for Tabular Data](https://mlsanity.com/paper/2310.13196/) (2023)
- [TabGenie: A Toolkit for Table-to-Text Generation](https://mlsanity.com/paper/2302.14169/) (2023)
- [Embedding Individual Table Columns for Resilient SQL Chatbots](https://mlsanity.com/paper/1811.00633/) (2018)

## Built on by (most cited first, 6 of 6)

- [Columbo: Expanding Abbreviated Column Names for Tabular Data Using Large Language Models](https://mlsanity.com/paper/2508.09403/) (2025)
- [TACO: Task-Aware Column Description Generation Using LLMs](https://mlsanity.com/paper/2606.21685/) (2026)
- [Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation](https://mlsanity.com/paper/2607.18029/) (2026)
- [ISEE: Interactive Semantic Enrichment for Database Fields](https://mlsanity.com/paper/2608.02604/) (2026)
- [Polaris: Learning to Generate Table Descriptions from Retrieval Feedback](https://mlsanity.com/paper/2608.17171/) (2026)
- [BudgetSchemaBench: A Budget-Swept Diagnostic for Schema Context in Text-to-SQL](https://mlsanity.com/paper/2610.00092/) (2026)

## Abstract (arXiv)

Relational databases often suffer from uninformative descriptors of table contents, such as ambiguous columns and hard-to-interpret values, impacting both human users and text-to-SQL models. In this paper, we explore the use of large language models (LLMs) to automatically generate detailed natural language descriptions for SQL database columns, aiming to improve text-to-SQL performance and automate metadata creation. We create a dataset of gold column descriptions based on the BIRD-Bench benchmark, manually refining its column descriptions and creating a taxonomy for categorizing column difficulty. We then evaluate several different LLMs in generating column descriptions across the columns and different difficulties in the dataset, finding that models unsurprisingly struggle with columns that exhibit inherent ambiguity, highlighting the need for manual expert input. We also find that incorporating such generated column descriptions consistently enhances text-to-SQL model performance, particularly for larger models like GPT-4o, Qwen2 72B and Mixtral 22Bx8. Notably, Qwen2-generated descriptions, containing by annotators deemed superfluous information, outperform manually curated gold descriptions, suggesting that models benefit from more detailed metadata than humans expect. Future work will investigate the specific features of these high-performing descriptions and explore other types of metadata, such as numerical reasoning and synonyms, to further improve text-to-SQL systems. The dataset, annotations and code will all be made available.