Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2401.02997 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909063335378944 |
|---|---|
| author | Domínguez, José Manuel Errázuriz, Benjamín Daher, Patricio |
| author_facet | Domínguez, José Manuel Errázuriz, Benjamín Daher, Patricio |
| contents | Large Language Models (LLMs) have gained considerable notoriety in the field of natural language to SQL tasks (NL2SQL). In this study, we show how task decomposition can greatly benefit LLMs in database understanding and query generation in order to answer human questions with an SQL query.
We fined-tuned open source models, specifically Llama-2 and Code Llama, by combining 2 different models each designated to focus on one of two tasks in order to leverage each model's core competency to further increase the accuracy of the final SQL query.
We propose a new framework to divide the schema into chunks in order to fit more information into a limited context. Our results are comparable with those obtained by GPT-4 at the same time being 135 times smaller, 90 times faster and more than 100 times cheaper than GPT-4. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_02997 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Blar-SQL: Faster, Stronger, Smaller NL2SQL Domínguez, José Manuel Errázuriz, Benjamín Daher, Patricio Computation and Language Artificial Intelligence Large Language Models (LLMs) have gained considerable notoriety in the field of natural language to SQL tasks (NL2SQL). In this study, we show how task decomposition can greatly benefit LLMs in database understanding and query generation in order to answer human questions with an SQL query. We fined-tuned open source models, specifically Llama-2 and Code Llama, by combining 2 different models each designated to focus on one of two tasks in order to leverage each model's core competency to further increase the accuracy of the final SQL query. We propose a new framework to divide the schema into chunks in order to fit more information into a limited context. Our results are comparable with those obtained by GPT-4 at the same time being 135 times smaller, 90 times faster and more than 100 times cheaper than GPT-4. |
| title | Blar-SQL: Faster, Stronger, Smaller NL2SQL |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2401.02997 |