AgentCourt: Simulating Court with Adversarial Evolvable Lawyer Agents

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Chen, Guhong, Fan, Liyang, Gong, Zihan, Xie, Nan, Li, Zixuan, Liu, Ziqiang, Li, Chengming, Qu, Qiang, Alinejad-Rokny, Hamid, Ni, Shiwen, Yang, Min
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912430737588224
author Chen, Guhong
Fan, Liyang
Gong, Zihan
Xie, Nan
Li, Zixuan
Liu, Ziqiang
Li, Chengming
Qu, Qiang
Alinejad-Rokny, Hamid
Ni, Shiwen
Yang, Min
author_facet Chen, Guhong
Fan, Liyang
Gong, Zihan
Xie, Nan
Li, Zixuan
Liu, Ziqiang
Li, Chengming
Qu, Qiang
Alinejad-Rokny, Hamid
Ni, Shiwen
Yang, Min
contents Current research in LLM-based simulation systems lacks comprehensive solutions for modeling real-world court proceedings, while existing legal language models struggle with dynamic courtroom interactions. We present AgentCourt, a comprehensive legal simulation framework that addresses these challenges through adversarial evolution of LLM-based agents. Our AgentCourt introduces a new adversarial evolutionary approach for agents called AdvEvol, which performs dynamic knowledge learning and evolution through structured adversarial interactions in a simulated courtroom program, breaking the limitations of the traditional reliance on static knowledge bases or manual annotations. By simulating 1,000 civil cases, we construct an evolving knowledge base that enhances the agents' legal reasoning abilities. The evolved lawyer agents demonstrated outstanding performance on our newly introduced CourtBench benchmark, achieving a 12.1% improvement in performance compared to the original lawyer agents. Evaluations by professional lawyers confirm the effectiveness of our approach across three critical dimensions: cognitive agility, professional knowledge, and logical rigor. Beyond outperforming specialized legal models in interactive reasoning tasks, our findings emphasize the importance of adversarial learning in legal AI and suggest promising directions for extending simulation-based legal reasoning to broader judicial and regulatory contexts. The project's code is available at: https://github.com/relic-yuexi/AgentCourt
format Preprint
id arxiv_https___arxiv_org_abs_2408_08089
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AgentCourt: Simulating Court with Adversarial Evolvable Lawyer Agents
Chen, Guhong
Fan, Liyang
Gong, Zihan
Xie, Nan
Li, Zixuan
Liu, Ziqiang
Li, Chengming
Qu, Qiang
Alinejad-Rokny, Hamid
Ni, Shiwen
Yang, Min
Computation and Language
Artificial Intelligence
Current research in LLM-based simulation systems lacks comprehensive solutions for modeling real-world court proceedings, while existing legal language models struggle with dynamic courtroom interactions. We present AgentCourt, a comprehensive legal simulation framework that addresses these challenges through adversarial evolution of LLM-based agents. Our AgentCourt introduces a new adversarial evolutionary approach for agents called AdvEvol, which performs dynamic knowledge learning and evolution through structured adversarial interactions in a simulated courtroom program, breaking the limitations of the traditional reliance on static knowledge bases or manual annotations. By simulating 1,000 civil cases, we construct an evolving knowledge base that enhances the agents' legal reasoning abilities. The evolved lawyer agents demonstrated outstanding performance on our newly introduced CourtBench benchmark, achieving a 12.1% improvement in performance compared to the original lawyer agents. Evaluations by professional lawyers confirm the effectiveness of our approach across three critical dimensions: cognitive agility, professional knowledge, and logical rigor. Beyond outperforming specialized legal models in interactive reasoning tasks, our findings emphasize the importance of adversarial learning in legal AI and suggest promising directions for extending simulation-based legal reasoning to broader judicial and regulatory contexts. The project's code is available at: https://github.com/relic-yuexi/AgentCourt
title AgentCourt: Simulating Court with Adversarial Evolvable Lawyer Agents
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2408.08089