JBE-QA: Japanese Bar Exam QA Dataset for Assessing Legal Domain Knowledge
This provides a comprehensive benchmark for assessing legal domain knowledge in Japanese LLMs, addressing a gap beyond prior resources focused only on Civil Code.
The authors tackled the problem of evaluating large language models' legal knowledge in Japanese by introducing JBE-QA, a dataset derived from the Japanese bar exam (2015-2024), covering Civil Code, Penal Code, and Constitution with 3,464 items, and found that proprietary models with reasoning performed best, with Constitution questions being easier.
We introduce JBE-QA, a Japanese Bar Exam Question-Answering dataset to evaluate large language models' legal knowledge. Derived from the multiple-choice (tanto-shiki) section of the Japanese bar exam (2015-2024), JBE-QA provides the first comprehensive benchmark for Japanese legal-domain evaluation of LLMs. It covers the Civil Code, the Penal Code, and the Constitution, extending beyond the Civil Code focus of prior Japanese resources. Each question is decomposed into independent true/false judgments with structured contextual fields. The dataset contains 3,464 items with balanced labels. We evaluate 26 LLMs, including proprietary, open-weight, Japanese-specialised, and reasoning models. Our results show that proprietary models with reasoning enabled perform best, and the Constitution questions are generally easier than the Civil Code or the Penal Code questions.