Description of doctoral student's silhouette

Witold Sosnowski specializes in artificial intelligence and natural language processing. His research focuses on developing artificial intelligence-based solutions to combat disinformation, with particular emphasis on three key areas: creating benchmarks to evaluate large language models, designing systems to analyze disinformation narratives, and building artificial intelligence-based models to detect and analyze such narratives.

He has actively participated in international research initiatives and collaborated with leading global institutions, including the Open University of Cyprus. His work plays a key role in improving the reliability and effectiveness of AI systems in identifying and mitigating misinformation, with significant impact on natural language processing, artificial intelligence ethics and the integrity of digital information.

Description of the subject matter of the research project

Vygotsky's research focuses on developing AI-powered methods for detecting and analyzing disinformation narratives using a structured, multi-faceted approach. The project begins with the creation of a benchmark to assess how well large-scale language models identify disinformation, both in its simpler form and under the influence of persuasive techniques such as Pathos, Ethos and Logos. The second part of the research aims to develop a system capable of detecting disinformation narratives in newspaper articles by comparing them to a predefined list of known disinformation patterns. Finally, the project aims to develop a system that can analyze large collections of texts to identify new disinformation narratives, which will provide a deeper understanding of current trends in disinformation. By combining these elements, Vygotsky's research seeks to improve the accuracy and reliability of AI-powered disinformation detection, contributing to a more comprehensive understanding of the spread and evolution of disinformation narratives in the media.

Selected publications