DTS-SQL: Decomposed Text-to-SQL with Small Large Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916112722034688 |
|---|---|
| author | Pourreza, Mohammadreza Rafiei, Davood |
| author_facet | Pourreza, Mohammadreza Rafiei, Davood |
| contents | Leading models for the text-to-SQL task heavily rely on proprietary Large Language Models (LLMs), posing concerns over data privacy. Closing the performance gap between small open-source models and large proprietary models is crucial to mitigate this reliance. To this end, we introduce a novel two-stage fine-tuning approach that decomposes the task into two simpler tasks. Through comprehensive evaluation on two large cross-domain datasets and two small LLMs, we show that this approach improves execution accuracy by 3 to 7 percent, effectively aligning the performance of open-source models with their proprietary counterparts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01117 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DTS-SQL: Decomposed Text-to-SQL with Small Large Language Models Pourreza, Mohammadreza Rafiei, Davood Computation and Language Databases Human-Computer Interaction Leading models for the text-to-SQL task heavily rely on proprietary Large Language Models (LLMs), posing concerns over data privacy. Closing the performance gap between small open-source models and large proprietary models is crucial to mitigate this reliance. To this end, we introduce a novel two-stage fine-tuning approach that decomposes the task into two simpler tasks. Through comprehensive evaluation on two large cross-domain datasets and two small LLMs, we show that this approach improves execution accuracy by 3 to 7 percent, effectively aligning the performance of open-source models with their proprietary counterparts. |
| title | DTS-SQL: Decomposed Text-to-SQL with Small Large Language Models |
| topic | Computation and Language Databases Human-Computer Interaction |
| url | https://arxiv.org/abs/2402.01117 |