Monte Carlo Tree Search with Reasoning Path Refinement for Small Language Models in Conversational Text-to-NoSQL

TL;DR

This paper introduces a novel framework called Stage-MCTS that enables small language models to generate accurate NoSQL queries in conversational settings by combining reasoning, search, and training strategies, supported by a new dataset.

Contribution

The paper presents Stage-MCTS, a reasoning and search-based approach for small language models to handle conversational text-to-NoSQL tasks, along with the CoNoSQL dataset for evaluation.

Findings

Outperforms state-of-the-art large models in execution value match accuracy.

Uses Monte Carlo Tree Search guided by rules for stepwise reasoning.

Achieves up to 7.93% improvement in accuracy.

Abstract

NoSQL databases have been widely adopted in big data analytics, geospatial applications, and healthcare services, due to their flexibility and scalability. However, querying NoSQL databases requires specialized technical expertise, creating a high barrier for users. While recent studies have explored text-to-NoSQL problem, they primarily focus on single-turn interactions, ignoring the conversational nature of real-world queries. To bridge this gap, we introduce the Conversational Text-to-NoSQL task, which generates NoSQL queries given a natural language question, a NoSQL database, and the dialogue history. To address this task, we propose Stage-MCTS, a framework that endows small language models (SLMs) with NoSQL-specific reasoning capabilities by formulating query generation as a search problem. The framework employs Monte Carlo Tree Search (MCTS) guided by a rule-based reward to produce stepwise reasoning data, followed by progressive supervised fine-tuning (SFT) and self-training strategies. We further construct CoNoSQL, a cross-domain dataset with over 2,000 dialogues and 150 databases, to support evaluation. Experiments demonstrate that our approach outperforms state-of-the-art large reasoning models, improving execution value match (EVM) accuracy by up to 7.93%.