Young Researchers at the Forefront of Scientific Applications of Artificial Intelligence
In recent months, several students participating in the AI Core Programme have been actively engaged in research projects exploring various applications of artificial intelligence. The overarching objective of these initiatives is to contribute to both the scientific understanding and practical utilization of modern AI technologies, while providing students with opportunities to participate in research activities of national and international relevance.
Among the ongoing projects, a particularly notable initiative is being conducted in collaboration with researchers from HUN-REN, focusing on the role of artificial intelligence in scientific research. One of the project's primary research directions involves the analysis of scientific publications using large language models. The researchers aim to investigate whether the future scientific impact and success of a research idea can be predicted based solely on the abstract of a scientific publication. In addition, the project examines the extent to which automated peer-review systems are capable of providing objective and reliable evaluations, as well as the potential biases that may emerge in comparison with traditional expert-led review processes.
Another significant area of research focuses on the analysis of AI-generated scientific hypotheses. As today's generative AI systems become increasingly capable of formulating complex scientific ideas and theories, an important question arises: do these outputs genuinely contribute novel scientific value? Within the project, researchers assess the originality of hypotheses generated by various artificial intelligence systems and examine the performance differences among different models. The research also investigates whether evaluation models trained on human-authored scientific publications can accurately and consistently assess the quality of scientific concepts generated by artificial intelligence.
Previous projects within the research portfolio of AI Core participants have also explored the application of Retrieval-Augmented Generation (RAG) technologies and large language models. These studies primarily focused on how state-of-the-art AI systems are able to retrieve, integrate, and utilize up-to-date information during response generation. Given the rapid pace of development in the field, research priorities continue to evolve, and these projects may expand to include new directions and emerging topics in the near future.
In another research initiative, students investigated and compared different learning-based artificial intelligence systems. The project involved analyzing the operational principles, learning mechanisms, and practical applicability of multiple AI-driven solutions. The findings highlighted significant opportunities for further research, particularly in the area of adaptive and self-improving systems. As a result of ongoing professional consultations, plans have also emerged to further integrate these students into projects associated with the HUN-REN Research Network, where they will have the opportunity to explore additional questions related to the scientific applications of artificial intelligence.
Currently, the AI Core Programme supports the research activities of four students through active research contracts. The ongoing projects clearly demonstrate that artificial intelligence is no longer viewed solely as an applied technological tool within the university environment; it is increasingly emerging as an independent field of scientific inquiry. The students’ work contributes both to a deeper understanding of contemporary AI solutions and to ensuring that the University and its research partners actively participate in the scientific discourse shaping the future of artificial intelligence.
One of the programme’s most significant achievements is that it enables young researchers to engage in high-level research projects during their studies, address questions of international relevance, and gain first-hand experience of the scientific innovation process. Based on the current projects and their results, the study of scientific applications of artificial intelligence is expected to remain one of the defining areas of the AI Core Programme in the years ahead.
Another significant area of research focuses on the analysis of AI-generated scientific hypotheses. As today's generative AI systems become increasingly capable of formulating complex scientific ideas and theories, an important question arises: do these outputs genuinely contribute novel scientific value? Within the project, researchers assess the originality of hypotheses generated by various artificial intelligence systems and examine the performance differences among different models. The research also investigates whether evaluation models trained on human-authored scientific publications can accurately and consistently assess the quality of scientific concepts generated by artificial intelligence.
Previous projects within the research portfolio of AI Core participants have also explored the application of Retrieval-Augmented Generation (RAG) technologies and large language models. These studies primarily focused on how state-of-the-art AI systems are able to retrieve, integrate, and utilize up-to-date information during response generation. Given the rapid pace of development in the field, research priorities continue to evolve, and these projects may expand to include new directions and emerging topics in the near future.
In another research initiative, students investigated and compared different learning-based artificial intelligence systems. The project involved analyzing the operational principles, learning mechanisms, and practical applicability of multiple AI-driven solutions. The findings highlighted significant opportunities for further research, particularly in the area of adaptive and self-improving systems. As a result of ongoing professional consultations, plans have also emerged to further integrate these students into projects associated with the HUN-REN Research Network, where they will have the opportunity to explore additional questions related to the scientific applications of artificial intelligence.
Currently, the AI Core Programme supports the research activities of four students through active research contracts. The ongoing projects clearly demonstrate that artificial intelligence is no longer viewed solely as an applied technological tool within the university environment; it is increasingly emerging as an independent field of scientific inquiry. The students’ work contributes both to a deeper understanding of contemporary AI solutions and to ensuring that the University and its research partners actively participate in the scientific discourse shaping the future of artificial intelligence.
One of the programme’s most significant achievements is that it enables young researchers to engage in high-level research projects during their studies, address questions of international relevance, and gain first-hand experience of the scientific innovation process. Based on the current projects and their results, the study of scientific applications of artificial intelligence is expected to remain one of the defining areas of the AI Core Programme in the years ahead.
