$R^3$: "This is My SQL, Are You With Me?" A Consensus-Based Multi-Agent System for Text-to-SQL Tasks
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914866023890944 |
|---|---|
| author | Xia, Hanchen Jiang, Feng Deng, Naihao Wang, Cunxiang Zhao, Guojiang Mihalcea, Rada Zhang, Yue |
| author_facet | Xia, Hanchen Jiang, Feng Deng, Naihao Wang, Cunxiang Zhao, Guojiang Mihalcea, Rada Zhang, Yue |
| contents | Large Language Models (LLMs) have demonstrated strong performance on various tasks. To unleash their power on the Text-to-SQL task, we propose $R^3$ (Review-Rebuttal-Revision), a consensus-based multi-agent system for Text-to-SQL tasks. $R^3$ outperforms the existing single LLM Text-to-SQL systems as well as the multi-agent Text-to-SQL systems by $1.3\%$ to $8.1\%$ on Spider and Bird. Surprisingly, we find that for Llama-3-8B, $R^3$ outperforms chain-of-thought prompting by over 20\%, even outperforming GPT-3.5 on the development set of Spider. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_14851 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | $R^3$: "This is My SQL, Are You With Me?" A Consensus-Based Multi-Agent System for Text-to-SQL Tasks Xia, Hanchen Jiang, Feng Deng, Naihao Wang, Cunxiang Zhao, Guojiang Mihalcea, Rada Zhang, Yue Computation and Language Artificial Intelligence Databases Large Language Models (LLMs) have demonstrated strong performance on various tasks. To unleash their power on the Text-to-SQL task, we propose $R^3$ (Review-Rebuttal-Revision), a consensus-based multi-agent system for Text-to-SQL tasks. $R^3$ outperforms the existing single LLM Text-to-SQL systems as well as the multi-agent Text-to-SQL systems by $1.3\%$ to $8.1\%$ on Spider and Bird. Surprisingly, we find that for Llama-3-8B, $R^3$ outperforms chain-of-thought prompting by over 20\%, even outperforming GPT-3.5 on the development set of Spider. |
| title | $R^3$: "This is My SQL, Are You With Me?" A Consensus-Based Multi-Agent System for Text-to-SQL Tasks |
| topic | Computation and Language Artificial Intelligence Databases |
| url | https://arxiv.org/abs/2402.14851 |