Overview of IberLEF 2024: Natural Language Processing Challenges for Spanish and other Iberian Languages

IberLEF is a shared evaluation campaign for Natural Language Processing systems focused on Spanish and other Iberian languages, organized annually since 2019 as part of the conference for the Spanish Society for Natural Language Processing. Its aim is to inspire the research community to develop and participate in competitive tasks related to text processing, understanding, and generation. These efforts are geared towards defining new research challenges and setting state-of-the-art results in Iberian languages, including Spanish, Portuguese, Catalan, Basque, and Galician. This paper provides an overview of the evaluation activities conducted during IberLEF 2024, which featured 12 tasks and 25 subtasks. These tasks covered various areas such as automatic text generation identification, biomedical Natural Language Processing, counter-speech, early risk prediction on the Internet, harmful and inclusive content detection, language reliability, political ideology and propaganda identification, and sentiment and emotion analysis. Overall, the IberLEF 2024 activities represented a significant collaborative effort, involving 289 researchers from 23 countries across Europe, Asia, Africa, Australia, and the Americas.