ACL 2025
Maria Poiaganova attended the 63rd Annual Meeting of the Association for Computational Linguistics in Vienna.
Maria presented her research in argument mining, a subfield of NLP focused on analyzing the structure of arguments in text. In the study, she examined how well language models can transfer their understanding of political arguments from one genre to another, training the models on U.S. presidential debates and testing them on United Nations Security Council speeches. The results suggest that identifying components of an argument is relatively straightforward even across genres, whereas capturing the relationships between these components remains a significant challenge.
You can read the full publication here.
Congratulations on this insightful research!