Text-to-SQL as Dual-State Reasoning: Integrating Adaptive Context and Progressive Generation

TL;DR

This paper introduces DSR-SQL, a dual-state reasoning framework for Text-to-SQL tasks that enhances reasoning coherence and schema understanding by modeling interaction between adaptive context and progressive generation, achieving competitive results.

Contribution

The paper presents a novel dual-state reasoning approach that improves Text-to-SQL performance by refining schema context and enabling feedback-guided SQL generation without additional training.

Findings

Achieves 35.28% execution accuracy on Spider 2.0-Snow

Attains 68.32% accuracy on BIRD development set

Does not require post-training or in-context examples

Abstract

Recent divide-and-conquer reasoning approaches, particularly those based on Chain-of-Thought (CoT), have substantially improved the Text-to-SQL capabilities of Large Language Models (LLMs). However, when applied to complex enterprise databases, such methods struggle to maintain coherent reasoning due to limited context capacity, unreliable schema linking, and weak grounding in database semantics. To overcome these issues, we introduce DSR-SQL, a \textbf{D}ual-\textbf{S}tate \textbf{R}easoning framework that models Text-to-SQL as an interaction between an adaptive context state and a progressive generation state. The first constructs a compact, semantically faithful environment by refining large schemas and selecting relevant structures, while the second formalizes SQL synthesis as feedback-guided state transitions, enabling the model to self-correct and align with user intent. Without any post-training or in-context examples, DSR-SQL achieves competitive performance, reaching 35.28\% execution accuracy on Spider 2.0-Snow and 68.32\% on BIRD development set. Our implementation will be open-sourced at: https://github.com/DMIRLAB-Group/DSR-SQL.