SafeQL: Search-based Refinement for Safe and Efficient LLM-based Text-to-SQL
2026-08-10 • Databases
DatabasesArtificial Intelligence
AI summaryⓘ
The authors address a problem where large language models sometimes create incorrect database queries that don't work because they reference wrong tables or columns. They propose SafeQL, which uses the database system itself to actively guide the fixing process by only correcting the broken parts of a query instead of rewriting the whole thing. SafeQL tests potential fixes step-by-step to find a query that runs correctly, making the process more accurate and efficient. Their experiments show this method works better than previous approaches that generate new queries from scratch after errors.
Large Language ModelsText-to-SQLDatabase SchemaSQL QueryDatabase Management SystemQuery RefinementError FeedbackSearch-Based RepairSpider BenchmarkBird Benchmark
Authors
Geonho Lee, Min-Soo Kim
Abstract
Large language models (LLMs) have advanced Text-to-SQL by enabling natural language interfaces to databases without task-specific fine-tuning. However, existing LLM-based systems remain unreliable, often generating SQL queries that are invalid under the database schema, referencing non-existent tables, attributes, functions, or values. Such errors persist because interactions with the database management system (DBMS) are typically limited to error messages, leaving it in a largely passive role during query refinement. This paper proposes SafeQL, \textit{a search-based refinement paradigm that redefines the role of the DBMS as an active guide in the refinement process}. Instead of regenerating entire queries after execution failure, SafeQL interprets DBMS feedback to incrementally repair only the erroneous components. Each refinement step is formulated as a guided search within a \textit{safe query space}, where candidate queries are progressively validated through DBMS execution, thereby converging to an executable query and preventing repeated regeneration of errors. Experiments on the Bird and Spider benchmarks show that SafeQL significantly improves execution accuracy and efficiency compared to regeneration-based methods.