Hybrid AI Models for Materials Discovery with Versioned Laboratory Datasets: A Cross-Site Validation Agenda

Abstract

Materials-screening datasets often mix experimental batches, computed descriptors and inconsistent processing metadata. The objective is to examine whether hybrid prediction models remain reliable when material families and laboratory batches are held out. Relevant methodological and data-quality literature informs the design of the sampling and validation strategy [3–7]. The proposed method is a reproducibility-focused benchmark built from versioned materials datasets and documented processing histories. Comparators will include physical-descriptor baselines, purely learned models and hybrid models combining domain constraints with machine learning. Related enterprise-storage studies inform the assessment of operational reliability [1] and the deployment assumptions adopted for system-level comparison [2]. The cross-site extension holds out an entire institution or deployment location rather than relying only on random test splits. Statistical and computational sources help define the reproducibility checks and comparison procedures [8–12]. The primary outcomes are out-of-family prediction error, calibration and stability of candidate rankings under data revisions. Potential biases include recipe duplicates, provenance gaps, leakage across related materials and sensitivity to batch-specific artefacts. Between-site heterogeneity would be examined through prespecified subgroup estimates and transparent domain-shift diagnostics. Additional research informs the prespecified stress tests and limits on interpretation [13–17]. The intended contribution is an auditable research protocol rather than an unsupported claim of effectiveness; related work frames its transferability limits [18–23].