Chatlaw: A Multi-Agent Legal Assistant based on a Role-Aligned Mixture-of-Experts Architecture
Chatlaw: многоагентный юридический ассистент на основе архитектуры смеси экспертов, согласованной с ролями
2026-05-01
SCID: 54.1/we94qjuc
Discuss with AI
Chinese legal systemLawBench benchmarkRole-Aligned Mixture-of-ExpertsUnified Qualification Exammulti-agent legal assistant
Figures from the paper
Abstract (AI)
Artificial Intelligence (AI) holds great potential in legal services, yet Large Language Models (LLMs) face two major challenges: limited knowledge of the Chinese legal system and vulnerability to hallucinations. To address these issues, we present Chatlaw, a multi-agent legal assistant. Chatlaw’s framework is designed to emulate the Standard Operating Procedures (SOP) of real law firms, where different roles (e.g., assistant, researcher, senior lawyer) collaborate on a case. To computationally mirror this collaborative structure, we developed a novel Role-Aligned Mixture-of-Experts (RA-MoE) architecture. In this system, the internal "experts" are specifically trained to align with the distinct tasks of each agent role (e.g., inquiry, analysis, drafting). These specialized agents (Legal Assistant, Researcher, etc.) then form the collaborative framework. When they interact with users, retrieve legal knowledge, analyze case details, or generate reliable consultations, the RA-MoE architecture intelligently routes their computations to the corresponding dedicated expert, ensuring each step is handled by the most qualified parameters. In evaluations, Chatlaw surpasses general-purpose AI models, including GPT-4, achieving a 7.73% improvement in accuracy on the LawBench benchmark and an 11-point higher score on the Unified Qualification Exam for Legal Professionals. Real-case studies and expert assessments further confirm its robustness. Chatlaw enhances the accessibility and reliability of legal services, advancing the provision of legal support to the public. Chatlaw is an advanced AI legal assistant that improves the accuracy and reliability of legal services. It integrates a high-quality legal dataset, a Role-Aligned Mixture-of-Experts (RA-MoE) architecture, and a multi-agent system to provide precise, case-specific legal advice. By emulating real law firm workflows, it reduces errors and outperforms general AI models like GPT-4, achieving better results in legal benchmarks and real-world evaluations. Chatlaw makes legal services more accessible, marking a significant step forward in AI for law.
Key Findings
1
Chatlaw improves accuracy by 7.73% on the LawBench benchmark and scores 11 points higher on the Unified Qualification Exam than general-purpose AI models, including GPT-4.
2
Chatlaw integrates a high-quality legal dataset and legal-knowledge retrieval to address limited Chinese legal knowledge and reduce hallucination-related errors.
3
Chatlaw introduces a multi-agent legal assistant that emulates law-firm standard operating procedures through collaborating role-specific agents.
4
Its Role-Aligned Mixture-of-Experts architecture routes inquiry, analysis, and drafting tasks to experts specialized for corresponding legal roles.
5
Real-case studies and expert assessments support Chatlaw’s robustness, accuracy, and reliability for case-specific legal consultations.
Research Object
Chatlaw, a multi-agent AI legal assistant for Chinese legal services
Research Subject
the accuracy, reliability, and robustness of case-specific legal consultations produced through role-aligned multi-agent collaboration and mixture-of-experts routing
Publication Details
Publication Date
2026-05-01
Journal
Publisher
ISSN
Cited by
0
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest