SOURCE-LINKED INTELLIGENCE

Neuro-Symbolic Agentic AI Research Assistant

CORDIS · observation · Publication date unknown

Neuro-Symbolic Agentic AI Research Assistant This PoC transforms our Neuro-Symbolic Science Knowledge Graph research into a deployable Research AIssistant that fuses large language models with the Open Research Knowledge Graph (ORKG). The assistant delivers knowledge-augmented generation grounded in structured, citable research contributions, with mandatory citations, constraint checks for study and method semantics, and uncertainty signalling. Building on results from the ERC-funded ScienceGraph project—now embodied in the ORKG and ORKG-ASK services—the PoC closes the loop from assisted synthesis back to curated knowledge by publishing vetted outputs as reusable, DOI-traceable artifacts. The PoC targets two real-world pilots: (A) Materials science, focusing on protocol advice and hypothesis comparison over process parameters and metrology for Atomic Layer Disposition and Etching processes; and (B) Data science in cooperation with NFDI4DataScience, emphasizing reproducibility signals across methods, datasets, and evaluation pipelines. Success crite

Read original source ↗ Open in workspace

recordType

award

status

SIGNED

region

EU

value

150000

unit

EUR