Overview of PAN 2024: Multi-author Writing Style Analysis, Multilingual Text Detoxification, Oppositional Thinking Analysis, and Generative AI Authorship Verification - Extended Abstract

Janek Bevendorff, Xavier Bonet Casals, Berta Chulvi, Daryna Dementieva, Ashaf Elnagar, Dayne Freitag, Maik Fröbe, Damir Korencic, Maximilian Mayerl, Animesh Mukherjee, Alexander Panchenko, Martin Potthast, Francisco Rangel, Paolo Rosso, Alisa Smirnova, Efstathios Stamatatos, Benno Stein, Mariona Taulé, Dmitry Ustalov, Matti Wiegmann et al. (1 additional authors not shown)

Abstract: The paper gives a brief overview of the four shared tasks organized at the PAN 2024 lab on digital text forensics and stylometry to be hosted at CLEF 2024. The goal of the PAN lab is to advance the state-of-the-art in text forensics and stylometry through an objective evaluation of new and established methods on new benchmark datasets. Our four tasks are: (1) multi-author writing style analysis, which we continue from 2023 in a more difficult version, (2) multilingual text detoxification, a new task that aims to translate and re-formulate text in a non-toxic way, (3) oppositional thinking analysis, a new task that aims to discriminate critical thinking from conspiracy narratives and identify their core actors, and (4) generative AI authorship verification, which formulates the detection of AI-generated text as an authorship problem, one of PAN’s core tasks. As with the previous editions, PAN invites software submissions as easy-to-reproduce docker containers; more than 400 pieces of software have been submitted from PAN’12 through PAN’23 combined, with all recent evaluations running on the TIRA experimentation platform [8].