JBE-QA: Japanese Bar Exam QA Dataset for Assessing Legal Domain Knowledge

TL;DR

JBE-QA is a new Japanese legal question-answering dataset derived from bar exam questions, designed to evaluate large language models' legal knowledge across multiple legal domains with structured, balanced items.

Contribution

It introduces the first comprehensive Japanese legal domain benchmark dataset for LLM evaluation, covering multiple legal codes and including structured question formats.

Findings

Proprietary models with reasoning perform best

Constitution questions are easier than Civil or Penal Code questions

The dataset contains 3,464 balanced items across legal domains

Abstract

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.