News

On Friday, March 27, a group of 17 undergraduate students from the Department of Humanities of the Democritus University of Thrace visited the Athena Research Center. The aim of the visit was to familiarize students with research and applications in the field of language technology.

Dr. Stella Markantonatou, Research Director at the Institute for Language and Speech Processing (ILSP) of the Athena Research Center, welcomed the students and provided a brief overview of the Center and its activities, with particular emphasis on the work of ILSP.

 

As part of the visit, an introduction to the technology behind Large Language Models was presented, along with an overview of the Natural Language Processing team of the Archimedes Unit of the Athena Research Center by PhD candidate and research associate George Moschovis. Particular emphasis was placed on biomedical question-answering systems and their evaluation methods, as well as on the collaboration between researchers of the Archimedes Unit and the Papageorgiou General Hospital in Thessaloniki for the development of systems for the automatic generation of diagnoses from echocardiography images and videos.

 

Subsequently, Stavros Bobolas, postdoctoral researcher at the Center, introduced the dialectal variation of Modern Greek and discussed Automatic Speech Recognition (ASR) models, focusing on adaptation methods for dialectal speech and on the challenges related to the creation and use of appropriate linguistic resources.

 

Next, Vivian Stamou, postdoctoral researcher at the Center, presented the process of morphosyntactic analysis using approaches based on knowledge transfer from higher-resource language varieties. She also highlighted the challenges that arise from specific linguistic features that either diverge from or converge with those of Standard Modern Greek, facilitating or complicating the analysis process.

 

Finally, PhD candidate and research associate Antonis Dimakis presented approaches for mapping dialectal texts to Standard Modern Greek, initially through rule-based methods and subsequently using Large Language Models, incorporating a small number of parallel sentence examples in both the dialect and the standard variety within the prompts.

 

During the visit, students gained insight into the strongly interdisciplinary nature of research at the Athena Research Center, across a wide range of applied research activities. The visit provided a meaningful opportunity to connect theoretical knowledge with practical applications, contributing to a better understanding of how scientific knowledge is transformed into practical tools and cutting-edge technologies.