LLM and Data Repositories

Presentation for the CLARIAH Tech and Data Day on the decentralized research data infrastructure, LLM and Data Repositories. We're building distributed knowledge graph based on Dataverse data network producing structured data and providing provenance information which could be used as ground truth for the Machine Learning in general and Large Language Models specifically. This approach also allows deep integration of LLM with KG providing researchers with opportunities to "talk" to data just like to an assistant.