Tag
This paper provides a comparative review of AI regulations across the EU, US, and China for high-risk use cases, analyzing FAIR principles and identifying governance gaps. It proposes Knowledge Blocks, a machine-checkable compliance artifact pattern to address implementation challenges.
This paper introduces Nutrition Data Service (NDS), source-preserving infrastructure that operationalizes FAIR principles for AI-agent-mediated nutrition research, addressing data identity, search, and crosswalk challenges. It demonstrates strong benchmark results and improved reproducibility compared to open-web reconstruction.
AgentFAIR is a multi-agent framework that uses LLM evaluators and a critic to assess FAIR compliance of geospatial datasets, achieving sub-principle agreement of 89% and Fleiss' κ=0.71 in expert studies, at a cost of $0.054 per dataset.
A novel framework called FAIR GraphRAG integrates FAIR Digital Objects with graph-based retrieval to enhance retrieval-augmented generation for semantic data analysis, improving question answering accuracy and adherence to FAIR principles, demonstrated on a biomedical dataset.