SteinerSQL: Graph-Guided Mathematical Reasoning for Text-to-SQL Generation
This addresses the challenge of generating accurate SQL queries from natural language for database users, presenting a new unified paradigm that is not incremental but offers a novel approach.
The paper tackled the problem of complex Text-to-SQL queries requiring mathematical reasoning and schema navigation by introducing SteinerSQL, a framework that unified these challenges into a graph-centric optimization problem, achieving state-of-the-art execution accuracies of 36.10% on LogicCat and 40.04% on Spider2.0-Lite benchmarks.
Large Language Models (LLMs) struggle with complex Text-to-SQL queries that demand both sophisticated mathematical reasoning and intricate schema navigation. Existing methods often tackle these challenges in isolation, creating a fractured reasoning process that compromises logical and structural correctness. To resolve this, we introduce SteinerSQL, a framework that unifies these dual challenges into a single, graph-centric optimization problem. SteinerSQL operates in three stages: mathematical decomposition to identify required tables (terminals), optimal reasoning scaffold construction via a Steiner tree problem, and multi-level validation to ensure correctness. On the challenging LogicCat and Spider2.0-Lite benchmarks, SteinerSQL establishes a new state-of-the-art with 36.10% and 40.04% execution accuracy, respectively, using Gemini-2.5-Pro. Beyond accuracy, SteinerSQL presents a new, unified paradigm for Text-to-SQL, paving the way for more robust and principled solutions to complex reasoning tasks.