Abstract: The study of harms in NLP is a fast-evolving field of research, which in a few years has seen the need of considering the subjectivity that characterizes this phenomenon. In this talk | present two complementary research projects that address this topic from two different perspectives. First, I discuss the systematic presence of bias against women and people with non-Western origin in data filtering strategies for harm reduction in pretraining datasets (Stranisci, & Hardmeier, C., 2025). Then, 1 describe the results of our study on canceling attitudes, whose perception appears to strongly rely on individuals’ moral stance rather than sociodemographic features (Lo, et al, 2025).

 

• Soda Marem Lo, Oscar Araque, Rajesh Sharma, and Marco Antonio Stranisci. 2025. That is Unacceptable: the Moral Foundations of Canceling. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6625-6639, Vienna, Austria. Association for Computational Linguistics.

• Stranisci, M. A., & Hardmeier, C. (2025). What Are They Filtering Out? A Survey of Filtering Strategies for Harm Reduction in Pretraining Datasets. arXiv preprint arXiv:2503.0572

 

Bio: Marco Antonio Stranisci is a postdoc researcher at the University of Turin. During his PhD, he worked at the intersection of Natural Language Processing and Semantic Web, focusing on the topic of bias in digital archives and on the interplay between hate speech, emotions, and moral values. In 2023 he founded aqua-tech, a start-up dedicated to designing inclusive and participatory NLP technologies.