ABOUT THE PROJECT

Language technologies are increasingly important in many areas of science. With the rapid advances in large-scale language models, they are positioning themselves as a core AI technology with far-reaching impact on scientific research, wider societal development, and society as a whole. Despite their great capabilities, large language models exhibit several shortcomings, such as inconsistent and incorrect answers, high computational demands, weak knowledge of less-resourced languages, poor performance in some important tasks and domains, and weaknesses in understanding society, ethics, and human needs. The project will improve the technology of large language models by enriching them with diverse knowledge and data, using them to improve the performance of speech technologies, and giving them access to reliable external information. Improvements in the core capabilities of large language models will lead to breakthroughs in several areas of the digital humanities: linguistics, lexicography, education, analysis of social phenomena and contemporary history, political sciences, folkloristics, and law.

The project brings together leading Slovenian researchers and institutions in language technologies. It connects them with prominent digital humanities researchers and institutions, fostering innovation and excellence. The project will organize numerous dissemination activities to bolster Slovenia’s international research stature and encourage public participation. 

Project Type: Gravitacija

Project code: GC-0002
Duration of the project: 1.10.2024 – 30.9.2027