用法律体系条件化大模型，检测德语仇恨言论的刑事定性
Conditioning Large Language Models on Legal Systems? Detecting Punishable Hate Speech
- 在宪法、成文法、判例法多层抽象下训练大模型
- 模型仍远低于法律专家水平，抽象知识导致自相矛盾
- 具体法律知识助于识别目标群体，但难判行为定性
法律问题的评估需结合特定法律体系及其多层级抽象（从宪法到成文法再到判例法）。当前大语言模型（LLMs）对法律体系的内化程度尚不明确。本文探讨在不同抽象层次上对大模型进行法律体系条件化的多种方法，聚焦于判断社交媒体内容是否构成德国刑法规定的煽动仇恨罪。结果表明，无论采用何种抽象层次的条件化，模型在仇恨言论的法律判断上与法律专家之间仍存在显著差距。分析发现，基于抽象法律知识的模型缺乏深层任务理解，常出现自我矛盾和幻觉；而使用具体法律知识的模型虽能较好识别目标群体，但在判定具体行为性质上表现不佳。
原文摘要 · Abstract (English)
The assessment of legal problems requires the consideration of a specific legal system and its levels of abstraction, from constitutional law to statutory law to case law. The extent to which Large Language Models (LLMs) internalize such legal systems is unknown. In this paper, we propose and investigate different approaches to condition LLMs at different levels of abstraction in legal systems. This paper examines different approaches to conditioning LLMs at multiple levels of abstraction in legal systems to detect potentially punishable hate speech. We focus on the task of classifying whether a specific social media posts falls under the criminal offense of incitement to hatred as prescribed by the German Criminal Code. The results show that there is still a significant performance gap between models and legal experts in the legal assessment of hate speech, regardless of the level of abstraction with which the models were conditioned. Our analysis revealed, that models conditioned on abstract legal knowledge lacked deep task understanding, often contradicting themselves and hallucinating answers, while models using concrete legal knowledge performed reasonably well in identifying relevant target groups, but struggled with classifying target conducts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。