Speaker: Yu Fan (PhD Candidate, Center for Law and Economics, ETH Zurich)
Long-form legal reasoning remains a key challenge for large language models (LLMs) in spite of recent advances in test-time scaling. We introduce LEXam, a novel benchmark derived from 340 law exams spanning 116 law school courses across a range of subjects and degree levels. The dataset comprises 4,886 law exam questions in English and German, including 2,841 long-form, open-ended questions and 2,045 multiple-choice questions. Besides reference answers, the open questions are also accompanied by explicit guidance outlining the expected legal reasoning approach such as issue spotting, rule recall, or rule application. Our evaluation on both open-ended and multiple-choice questions present significant challenges for current LLMs; in particular, they notably struggle with open questions that require structured, multi-step legal reasoning. Moreover, our results underscore the effectiveness of the dataset in differentiating between models with varying capabilities. Adopting an LLM-as-a-Judge paradigm with rigorous human expert validation, we demonstrate how model-generated reasoning steps can be evaluated consistently and accurately. Our evaluation setup provides a scalable method to assess legal reasoning quality beyond simple accuracy metrics.
Yu Fan is a PhD candidate at the Center for Law and Economics at ETH Zurich and an associated researcher at the ETH AI Center. He is supervised by Prof. Elliott Ash and co-supervised by Prof. Mrinmaya Sachan. Prior to joining the Center, he worked as a research associate at D-MTEC, ETH Zurich. His research interests include natural language processing, legal NLP, and computational social science, with current projects focusing on legal reasoning and retrieval.
