LitCurate: A Configuration-Driven AI-Assisted Framework for Scientific Database Construction with an Application to Lower-Mantle Equation-of-State Data
Information RetrievalArtificial Intelligence
Summary
The authors created LitCurate, a tool that helps turn lots of scientific papers into organized databases using smart language models. It carefully finds important studies, pulls out detailed information, and keeps track of where each fact came from, so scientists can check or fix parts anytime. They tested LitCurate by making a database about how certain deep-earth minerals behave under pressure, collecting data from over 200 papers. This database links detailed mineral info to its sources and is easy to search online. Overall, the authors show a way to make complex scientific knowledge easier to use for research and modeling.
Authors
Abin Shakya, Wilson Samuels, Dominica Wilson, Gioia A. Marchi, Israa Draz, Chenxing Luo, Renata M. Wentzcovitch
Abstract
The growing scientific literature contains decades of experimental and computational results that could support data-driven and physics-based modeling, yet much of this infor- mation remains locked in publications and is not readily usable for large-scale analysis or sci- entific software. Building structured databases from the literature is particularly challenging whenrelevantstudiesmustfirstbediscoveredamonglargecollectionsofpapersandreported quantities must be extracted with enough scientific context to remain usable. We present LitCurate, an open-source framework for building scientific databases from the literature using large language models within an auditable, stage-wise curation workflow. LitCurate integratesliteraturediscovery, relevancescreening, full-textprocessing, andstructuredinfor- mation extraction while retaining intermediate results and provenance, allowing researchers to inspect and revise individual stages rather than treating automated curation as a black- box process. We apply LitCurate to construct an equation-of-state database of lower-mantle and lower-mantle-relevant high-pressure mineral phases from experimental and theoretical studies, comprising 1,334 entries from 205 papers. The resulting dataset links reported equation-of-state parameters to mineral phases, compositions, equation formulations, meth- ods, and parameter constraints, and labels values as source-reported or citation-reported when provenance can be determined. The records are available through a searchable web application. By connecting scientific literature to traceable, machine-readable data, LitCu- rate provides a reusable approach for transforming accumulated literature into resources for scientific analysis and computational modeling.