Search

Feature engineering remains central to materials informatics, yet systematically introduces scientific blind spots that constrain discovery and interpretation. These blind spots arise from choices in descriptor selection, transformation, and dimensionality reduction that inadvertently prioritize statistical correlations over physical invariance, overlook multi-scale interactions, and embed dataset-specific biases into model architectures. In small-data regimes common to materials science, engineered features often amplify overfitting while diminishing generalizability across chemical spaces. Interpretability suffers as complex engineered descriptors obscure mechanistic linkages between atomic structure and macroscopic properties. Literature consistently highlights these limitations across perovskites, alloys, energy materials, and porous systems, underscoring the tension between predictive performance and scientific fidelity. This conceptual manuscript synthesizes these challenges and proposes an original Integrated Blind Spot Navigation Model (IBSNM). The framework organizes feature engineering around four interdependent pillars—physical consistency guardrails, multi-scale descriptor integration, uncertainty-aware selection, and iterative co-interpretation—linked by feedback mechanisms that surface and mitigate hidden assumptions. By reframing feature engineering as a navigable landscape rather than a static preprocessing step, the model offers a conceptual pathway toward more robust, transparent materials informatics practices that do not rely on empirical validation.

High-entropy alloys (HEAs) and multi-principal element alloys (MPEAs) occupy an enormous compositional space that conventional computational approaches cannot fully explore. A five-element system with 10 % concentration steps already contains millions of distinct nominal compositions, each further multiplied by exponentially many atomic configurations arising from configurational disorder. Machine learning (ML) interatomic potentials have been proposed as a scalable solution to accelerate property prediction and materials design in this vast space. Yet the compositional generalizability of these potentials—their capacity to make reliable predictions for compositions and local environments lying outside the training distribution—remains largely unexamined in a systematic way. Most existing benchmarks focus on interpolation within narrow ranges of ordered or equiatomic compounds, leaving critical extrapolation scenarios untested. This conceptual framework article identifies four distinct but interrelated dimensions of compositional generalizability in HEA ML potentials: (i) element extrapolation, (ii) concentration interpolation, (iii) multi-element recombination, and (iv) local environment diversity. It proposes a four-component assessment framework—training-set characterization, test-set design, dimension-specific generalization metrics, and a structured validation protocol—that enables researchers to quantify generalization gaps without relying on performance numbers or simulation results. For each dimension, explicit validation strategies and success criteria are defined, grounded in the literature on special quasirandom structures, graph-network representations, and equivariant architectures. The framework is deliberately conceptual, emphasizing definitions, relationships among components, assessment criteria, and diagnostic reasoning rather than empirical data. By adopting this framework, the community can move beyond ad-hoc testing and develop ML potentials that truly generalize across the high-dimensional alloy landscape. The implications extend to more reliable high-throughput screening, accelerated discovery of novel HEAs, and clearer guidance for training-set design and model architecture choices. Ultimately, systematic assessment of compositional generalizability will help ensure that ML potentials fulfill their promise for complex concentrated alloys.

Foundation models promise a transformative advance in materials science by enabling zero-shot property prediction across diverse chemistries, crystal structures, and physical properties, ostensibly eliminating the need for property-specific labelled datasets and expensive first-principles calculations. Proponents claim that models pre-trained on millions of structures can directly predict formation energies, band gaps, elastic moduli, thermal conductivities, and other attributes for entirely unseen materials without task-specific fine-tuning. This critique demonstrates that such claims largely reflect an illusion of generalizability. Reported zero-shot performance is systematically inflated by six pervasive artefacts: (1) test-set leakage of crystal structures or near-identical analogues from the pre-training corpus, (2) exploitation of strong inter-property correlations, (3) evaluation within the interpolation regime of the pre-training distribution, (4) benchmark bias inherent in widely reused datasets such as the Materials Project, (5) inconsistent pre-processing, data splits, and reporting practices, and (6) the absence of rigorous elementary baselines. Analysis of recent foundation-model literature (2023–2026) shows that zero-shot accuracies frequently collapse once these confounds are controlled and often fail to surpass simple statistical predictors such as k-nearest-neighbour regression on embeddings or linear models based on property correlations. The illusion is particularly acute in materials science due to the repeated reuse of identical structures across property databases and the dense network of physical correlations among computed quantities. Without stringent controls, overstated zero-shot claims risk misdirecting research resources, inflating expectations, and undermining trust in AI-driven materials discovery. This paper provides a diagnostic framework for identifying these artefacts, marshals supporting evidence from the contemporary literature, and proposes a six-point evaluation standard for credible zero-shot claims. Adoption of these practices is essential if foundation models are to deliver genuine advances in out-of-distribution generalization rather than repackaged statistical regularities.