Overview of AuTexTification at IberLEF 2023: Detection and Attribution of Machine-Generated Text in Multiple Domains

• DOI: 10.48550/arxiv.2309.11285
• Publication Date: 2023-01-01
• Authors: Areg Mikael Sarva...; José Ángel González; Marc Franco-Salvador; Francisco Rangel; Berta Chulvi; Paolo Rosso
• Keywords: Authorship Attribution

This paper presents the overview of AuTexTification shared task as part IberLEF 2023 Workshop in Iberian Languages Evaluation Forum, within framework SEPLN conference. consists two subtasks: for Subtask 1, participants had to determine whether a text is human-authored or has been generated by large language model. For 2, attribute machine-generated one six different generation models. Our dataset contains more than 160.000 texts across languages (English and Spanish) five domains (tweets, reviews, news, legal, how-to articles). A total 114 teams signed up participate, which 36 sent 175 runs, 20 them their working notes. In this overview, we present task, submitted participating systems, results.