Multi-Agent Simulator Drives Language Models for Legal Intensive Interaction
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915145338322944 |
|---|---|
| author | Yue, Shengbin Huang, Ting Jia, Zheng Wang, Siyuan Liu, Shujun Song, Yun Huang, Xuanjing Wei, Zhongyu |
| author_facet | Yue, Shengbin Huang, Ting Jia, Zheng Wang, Siyuan Liu, Shujun Song, Yun Huang, Xuanjing Wei, Zhongyu |
| contents | Large Language Models (LLMs) have significantly advanced legal intelligence, but the scarcity of scenario data impedes the progress toward interactive legal scenarios. This paper introduces a Multi-agent Legal Simulation Driver (MASER) to scalably generate synthetic data by simulating interactive legal scenarios. Leveraging real-legal case sources, MASER ensures the consistency of legal attributes between participants and introduces a supervisory mechanism to align participants' characters and behaviors as well as addressing distractions. A Multi-stage Interactive Legal Evaluation (MILE) benchmark is further constructed to evaluate LLMs' performance in dynamic legal scenarios. Extensive experiments confirm the effectiveness of our framework. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_06882 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-Agent Simulator Drives Language Models for Legal Intensive Interaction Yue, Shengbin Huang, Ting Jia, Zheng Wang, Siyuan Liu, Shujun Song, Yun Huang, Xuanjing Wei, Zhongyu Computation and Language Artificial Intelligence Large Language Models (LLMs) have significantly advanced legal intelligence, but the scarcity of scenario data impedes the progress toward interactive legal scenarios. This paper introduces a Multi-agent Legal Simulation Driver (MASER) to scalably generate synthetic data by simulating interactive legal scenarios. Leveraging real-legal case sources, MASER ensures the consistency of legal attributes between participants and introduces a supervisory mechanism to align participants' characters and behaviors as well as addressing distractions. A Multi-stage Interactive Legal Evaluation (MILE) benchmark is further constructed to evaluate LLMs' performance in dynamic legal scenarios. Extensive experiments confirm the effectiveness of our framework. |
| title | Multi-Agent Simulator Drives Language Models for Legal Intensive Interaction |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2502.06882 |