We present a new annotated corpus for the task of Nested Entity Recognition from legal acts in the EU acquis, the collection of common rights and obligations that constitute the body of EU law and is incorporated into the legal systems of EU Member States. The corpus is subdivided into two subcorpora: GOLD - human annotated (41 documents) and SILVER - created semi-automatically (537 documents). They contain 28.5K and 275K entity mentions labelled with fine-grained entity types using a dedicated taxonomy of 45 fine-grained entity types. We describe the corpus creation process and present corpus statistics. We use knowledge-, transition-, transformer-, and generative AI-based models to provide baselines and benchmark results.

We report on experiments on information extraction (IE) from EU Acquis, the European Union law. We introduce a new IE task of Information Provision Activity Requirement Extraction. This task comprises the identification of text fragments that introduce an obligation to provide information, and the extraction of structured information about the key entities involved along with the temporal modalities. We compare various technologies for this task, i.e. knowledge-, classical ML-, transformer-, and generative AI-based approaches, on a new benchmark corpus.