The International Workshop on Computational Methods for Online Discourse Analysis (BeyondFacts) is an annual workshop collocated with the 34th ACM International Web Conference 2025. BeyondFacts brings together scholars from communication studies, computational linguistics, and computer science to tackle the complexities of online discourse. Topics include argument mining, claim detection, stance analysis, and computational fact-checking. The workshop emphasizes the need for shared definitions and interdisciplinary collaboration to model and interpret claims, stances, and doubts across diverse contexts.

From April 28 to May 2, 2025, the 5th BeyondFacts conference was held at Sydney, Australia. This year we presented one of our studies entitled “Developing a Stance-induced Epidemiological Model to Examine Polarized Information Contagion.” The study introduced a stance-induced epidemiological model (SEIAIDZ)—a model that segments a population into sections depending on whether a user is susceptible, exposed, infected, or skeptical to information campaigns—that differentiated between users who agreed or disagreed with information. The model outperformed traditional contagion models that did not consider the polarized nature of the discourse. By incorporating content stance rather than just metadata, the model identified critical factors that help in dynamically measuring the information spread.