The Measurement Gap in AI and Doctrinal Reasoning in EU Law

The integration of large language models (LLMs) into legal practice raises significant questions regarding their ability to perform doctrinal legal reasoning. The current benchmarks fail to measure th