LEXam Legal Reasoning Benchmark Dataset

*This dataset supports online use.Click here to jump.

LEXam is a multilingual legal reasoning benchmark dataset jointly released by researchers from ETH Zurich, the Swiss Federal Supreme Court, the Max Planck Institute, and the University of Zurich. The relevant paper results are:LEXam: Benchmarking Legal Reasoning on 340 Law Exams".

The dataset contains 340 real legal examinations from different courses and different levels (undergraduate and master's) from the Faculty of Law of the University of Zurich, Switzerland, covering Swiss, European and international law, as well as legal theory and legal history.

The dataset contains 4886 questions, including:

• 2,841 long-form open-ended questions
• 2045 Multiple Choice Questions (MCQs)

These questions are written in English and German respectively, combining the diverse legal cultural backgrounds of civil law (emphasis on statutes and codes) and common law (emphasis on case law).