Babelscape/REDFM

Relation Extraction (RE) is a task that identifies relationships between entities in a text, enabling the acquisition of relational facts and bridging the gap between natural language and structured knowledge. However, current RE models often rely on small datasets with low coverage of relation types, particularly when working with languages other than English. \In this paper, we address the above issue and provide two new resources that enable the training and evaluation of multilingual RE systems. First, we present SRED\textsuperscript{FM}, an automatically annotated dataset covering 18 languages, 400 relation types, 13 entity types, totaling more than 40 million triplet instances. Second, we propose RED\textsuperscript{FM}, a smaller, human-revised dataset for seven languages that allows for the evaluation of multilingual RE systems. To demonstrate the utility of these novel datasets, we experiment with the first end-to-end multilingual RE model, mREBEL, that extracts triplets, including entity types, in multiple languages. We release our resources and model checkpoints at \href{https://www.github.com/babelscape/rebel}{https://www.github.com/babelscape/rebel}.