Multi-Legal-Bench: Evaluating LLMs on Legal Reasoning Across Jurisdictions, Languages, and Legal Traditions
Multi-Legal-Bench: 跨司法管辖区、语言和法律传统的法律推理评估LLM
机构 * SecondLayer
专题命中 安全评测 :alignment(abstract);分类 cs.CL、cs.AI
AI总结 提出Multi-Legal-Bench,首个跨司法管辖区法律基准,在6个国家、4个语系和1.34亿份法院判决上评估LLM,发现少样本效果跨辖区复制、无单一模型主导所有语言、跨语言迁移不遵循语言邻近性、分词器效率不显著预测跨语言准确率。
Comments 17 pages, 5 figures, 9 tables. v2 corrects scorer and taxonomy defects, adds no-model baselines showing label leakage, re-runs the Lithuanian cells on de-leaked text, and withdraws the claim that few-shot helps on judgment-form classification everywhere; all tables and figures regenerated. Dataset: https://huggingface.co/datasets/overthelex/multi-legal-bench