About the project
We think there's a gap in how CBDC projects get compared — and a dataset was the way to close it.
Why we're doing this
Trackers of central bank digital currencies are useful for one question — how far along is a project? — and mostly silent on the ones we think matter more: what is it for, what is it built on, and how does it treat privacy? We built this dataset to answer those questions from official or authority-attributed documents rather than tracker labels. The current v10.2e workflow uses two independent model providers, source-span verification, explicit ensemble rules and a probability-sampled double-blind human audit.
Work in progress. The independent blind audit is complete, while consensus adjudication and taxonomy refinement remain open. Future research will test improving LLMs on a larger and better-balanced document sample and with a more precisely operationalized family structure.
Team
Martin Bok
Department of International Economic Relations, Faculty of International Relations, Prague University of Economics and Business.
Dominik Stroukal
Metropolitan University Prague.
Contact
Questions, corrections, or collaboration ideas: use the contribute page, write to bokm00@vse.cz, or contact dominik.stroukal@mup.cz.