Speaker
Description
Recent advancements in AI, such as the availability of large language models (LLMs) and the ever-increasing amount of open access research papers, are creating new opportunities for improving how research is conducted, managed, organised, disseminated and used. This talk will first introduce CORE, a not-for-profit service that harvests openly available research papers from repositories, focusing on CORE’s ingestion pipeline and how papers in CORE can be accessed for machine processing.
Petr will then talk about a range of scenarios in which his team has been applying AI models over this scholarly content, including question-answering systems utilising LLMs (CORE-GPT), rapid systematic reviews and citation intent classification for research assessment, reflecting on the opportunities to improve research workflows and challenges in using AI responsibly. In concluding the talk with examples of using machine learning to organise research knowledge in repositories and libraries, Petr will argue that AI in research should not only be seen as the future but also as the present. A Q&A will follow the presentation.
