From Query to Counsel: Structured Reasoning with a Multi-Agent Framework and Dataset for Legal Consultation

TL;DR

This paper introduces JurisCQAD, a large Chinese legal query dataset, and JurisMA, a multi-agent framework for structured, context-aware legal question answering that outperforms existing models.

Contribution

It presents a novel dataset and a modular multi-agent system that enhances legal consultation accuracy through structured reasoning and interpretability.

Findings

JurisMA significantly outperforms general-purpose LLMs on LawBench.

The element graph improves context-aware reasoning in legal QA.

Structured decomposition enhances interpretability and accuracy.

Abstract

Legal consultation question answering (Legal CQA) presents unique challenges compared to traditional legal QA tasks, including the scarcity of high-quality training data, complex task composition, and strong contextual dependencies. To address these, we construct JurisCQAD, a large-scale dataset of over 43,000 real-world Chinese legal queries annotated with expert-validated positive and negative responses, and design a structured task decomposition that converts each query into a legal element graph integrating entities, events, intents, and legal issues. We further propose JurisMA, a modular multi-agent framework supporting dynamic routing, statutory grounding, and stylistic optimization. Combined with the element graph, the framework enables strong context-aware reasoning, effectively capturing dependencies across legal facts, norms, and procedural logic. Trained on JurisCQAD and evaluated on a refined LawBench, our system significantly outperforms both general-purpose and legal-domain LLMs across multiple lexical and semantic metrics, demonstrating the benefits of interpretable decomposition and modular collaboration in Legal CQA.