A universal LLM Framework for General Query Refinements

2026-02-17

AI Summary

This work proposes OmniTune, a novel framework that addresses the limitations of existing methods in SQL query tuning, which are often restricted to specific query types and simple constraints. OmniTune is the first approach capable of universally tuning arbitrary SQL queries under complex constraints. Built upon large language models (LLMs), it integrates optimization prompts (OPRO), a two-stage search strategy, historical summarization, and a skyline-based feedback mechanism to efficiently explore the tuning subspace and generate high-quality candidate configurations. Experimental results demonstrate that OmniTune not only subsumes the capabilities of current approaches on standard tasks but also significantly outperforms them in more complex tuning scenarios, exhibiting superior generality and effectiveness.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Numerous studies have explored the SQL query refinement problem, where the objective is to minimally modify an input query so that it satisfies a specified set of constraints. However, these works typically target restricted classes of queries or constraints. We present OmniTune, a general framework for refining arbitrary SQL queries using LLM-based optimization by prompting (OPRO). OmniTune employs a two-step OPRO scheme that explores promising refinement subspaces and samples candidates within them, supported by concise history and skyline summaries for effective feedback. Experiments on a comprehensive benchmark demonstrate that OmniTune handles both previously studied refinement tasks and more complex scenarios beyond the scope of existing solutions.

Problem

• SQL query refinement
• general framework
• query constraints
• arbitrary SQL queries
• optimization

Innovation

• LLM-based optimization
• OPRO
• SQL query refinement
• skyline summary
• general framework