Reproduce “From FAIR to AWARE: Data Readiness for Agentic AI in the Earth Sciences” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Ivan Kuznetsov, Dmitrii Pantiukhin, Sonja Hänzelmann, Fabian Wachsmann, Jonas Sølvsteen, Daniel Wiesmann, Thomas Jung, Nikolay Koldunov

The authors propose AWARE, an agent-specific layer extending FAIR, to make Earth-science data safer for autonomous reuse. It attaches machine-actionable handling instructions, scientific limits, identity, and provenance to data, alongside bounded validated tools and checks. In a seven-task hydrology study, constrained tool workflows achieved about 82% first-attempt success versus about 5% for generated code; this tests tool use, not AWARE itself.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.