ROUTE: Robust Multitask Tuning and Collaboration for Text-to-SQL
Despite the significant advancements in Text-to-SQL (Text2SQL) facilitated by large language models (LLMs), the latest state-of-the-art techniques are still trapped in the in-context learning of closed-source LLMs (e.g., GPT-4), which limits their applicability in open scenarios. To address this cha...
Gespeichert in:
Hauptverfasser: | , , , , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Despite the significant advancements in Text-to-SQL (Text2SQL) facilitated by
large language models (LLMs), the latest state-of-the-art techniques are still
trapped in the in-context learning of closed-source LLMs (e.g., GPT-4), which
limits their applicability in open scenarios. To address this challenge, we
propose a novel RObust mUltitask Tuning and collaboration mEthod (ROUTE) to
improve the comprehensive capabilities of open-source LLMs for Text2SQL,
thereby providing a more practical solution. Our approach begins with
multi-task supervised fine-tuning (SFT) using various synthetic training data
related to SQL generation. Unlike existing SFT-based Text2SQL methods, we
introduced several additional SFT tasks, including schema linking, noise
correction, and continuation writing. Engaging in a variety of SQL generation
tasks enhances the model's understanding of SQL syntax and improves its ability
to generate high-quality SQL queries. Additionally, inspired by the
collaborative modes of LLM agents, we introduce a Multitask Collaboration
Prompting (MCP) strategy. This strategy leverages collaboration across several
SQL-related tasks to reduce hallucinations during SQL generation, thereby
maximizing the potential of enhancing Text2SQL performance through explicit
multitask capabilities. Extensive experiments and in-depth analyses have been
performed on eight open-source LLMs and five widely-used benchmarks. The
results demonstrate that our proposal outperforms the latest Text2SQL methods
and yields leading performance. |
---|---|
DOI: | 10.48550/arxiv.2412.10138 |