no code implementations • SemEval (NAACL) 2022 • Xi Chen, Ali Zeynali, Chico Camargo, Fabian Flöck, Devin Gaffney, Przemyslaw Grabowicz, Scott Hale, David Jurgens, Mattia Samory
Thousands of new news articles appear daily in outlets in different languages.
no code implementations • 14 May 2024 • Zehui Yu, Indira Sen, Dennis Assenmacher, Mattia Samory, Leon Fröhling, Christina Dahn, Debora Nozza, Claudia Wagner
Yet, ML tools can only be as capable as the quality of the data they are trained on allows them.
1 code implementation • 1 May 2024 • Xi Chen, Scott A. Hale, David Jurgens, Mattia Samory, Ethan Zuckerman, Przemyslaw A. Grabowicz
News coverage profoundly affects how countries and individuals behave in international relations.
1 code implementation • 2 Nov 2023 • Indira Sen, Dennis Assenmacher, Mattia Samory, Isabelle Augenstein, Wil van der Aalst, Claudia Wagner
CADs introduce minimal changes to existing training data points and flip their labels; training on them may reduce model dependency on spurious features.
no code implementations • NAACL 2022 • Indira Sen, Mattia Samory, Claudia Wagner, Isabelle Augenstein
Especially, construct-driven CAD -- perturbations of core features -- may induce models to ignore the context in which core features are used.
no code implementations • 22 Apr 2022 • Shruti Phadke, Mattia Samory, Tanushree Mitra
In this paper, we provide the empirical modeling of various radicalization phases in online CT discussion participants.
no code implementations • 21 Jul 2021 • Shruti Phadke, Mattia Samory, Tanushree Mitra
We use these social imaginaries to create a computational framework for distinguishing belief and dissonance from general discussion about QAnon.
1 code implementation • 27 Apr 2020 • Mattia Samory, Indira Sen, Julian Kohne, Fabian Floeck, Claudia Wagner
By providing a scale-based codebook and insights regarding the shortcomings of the state-of-the-art, we hope to contribute to the development of better and broader models for sexism detection, including reflections on theory-driven approaches to data collection.