Marco A. Stranisci

Hi, I’m Marco: a postdoctoral researcher working at the intersection between Natural Language Processing and Semantic Web. In the past years I explored several topics that share the aim of designing fair and inclusive language technologies. My PhD thesis focused on the detection of allocative and representational bias in digital archives but I also worked on the psychological factors that explain the propagation of Hate Speech (e.g., coping strategies, moral configurations).

My actual interests revolve around two main directions: exploring the impact of pretraining in the propagation of harms by Large Language Models and adopting uncertainty as a proxy to detect social bias in models behavior. From a methodological perspective, I am interested in Neuro-symbolic applications and specifically in the integration of structured knowledge with NLP systems.

As a side quest, in 2023 I founded aequa-tech: a start-up whose aim is the development of NLP technologies for social good.