IESR:Efficient MCTS-Based Modular Reasoning for Text-to-SQL with Large Language Models

A framework called IESR (Information Enhanced Structured Reasoning) achieved state-of-the-art results on the complex reasoning benchmark LogicCat with 24.28 EX and on the Archer dataset with 37.28 EX using only compact lightweight large language models without fine-tuning, according to an arXiv paper (2602.05385v2). IESR combines LLM-based key information understanding and schema linking, decoupled mathematical computation and SQL generation, a multi-path Monte Carlo Tree Search (MCTS) reasoning mechanism with majority voting, and a trajectory consistency verification module with a discriminator model. The authors also report that current coder models show notable biases and deficiencies in physical knowledge, mathematical computation, and common-sense reasoning, and released code at https://github.com/Ffunkytao/IESR-SLM.

arXiv:2602.05385v2 Announce Type: replace Abstract: Text-to-SQL is a key natural language processing task that maps natural language questions to SQL queries, enabling intuitive interaction with web-based databases. Although current methods perform well on benchmarks like BIRD and Spider, they struggle with complex reasoning, domain knowledge, and hypothetical queries, and remain costly in enterprise deployment. To address these issues, we propose a framework named IESR Information Enhanced Structured Reasoning for lightweight large language models: i leverages LLMs for key information understanding and schema linking, and decoupling mathematical computation and SQL generation, ii integrates a multi-path reasoning mechanism based on Monte Carlo Tree Search MCTS with majority voting, and iii introduces a trajectory consistency verification module with a discriminator model to ensure accuracy and consistency. Experimental results demonstrate that IESR achieves state-of-the-art performance on the complex reasoning benchmark LogicCat 24.28 EX and the Archer dataset 37.28 EX using only compact lightweight models without fine-tuning. Furthermore, our analysis reveals that current coder models exhibit notable biases and deficiencies in physical knowledge, mathematical computation, and common-sense reasoning, highlighting important directions for future research. We released code at https://github.com/Ffunkytao/IESR-SLM.