Semantic Croissant and Cross-Domain Interoperability Framework (CDIF)

The Cross-Domain Interoperability Framework (CDIF) addresses one of the central challenges of modern science: enabling meaningful interoperability across disciplines, data infrastructures, and communities while preserving scientific context and trust. As data volumes grow and automation increases, interoperability cannot rely on technology alone - it must remain guided, validated, and governed by humans.

CDIF provides a semantic framework that connects datasets, metadata, variables, and concepts across domains using shared knowledge graph structures, mappings, and standards. Rather than enforcing a single global schema, CDIF supports domain autonomy while enabling alignment through common reference models, controlled vocabularies, and ontologies.

A key principle of CDIF is human-in-the-loop interoperability. Domain experts, data stewards, and curators play an active role in:

• Defining and validating semantic mappings across disciplines

• Interpreting complex or ambiguous concepts

• Ensuring provenance, trust, and scientific meaning are preserved

Automation with AI including semantic enrichment, recommendation, and validation, and used to augment human expertise, not replace it. This balanced approach enables scalable interoperability while maintaining accountability, inclusiveness, and scientific rigor.