This review aims to offer a conceptual framework and a solid reference for building intelligent multimodal material selection systems and presents a forward-looking research agenda covering self-supervised learning, knowledge-enhanced models, and interpretable human–AI collaboration.
Abstract
Data-driven material selection is progressively changing how materials are evaluated in engineering, manufacturing, and product design. With the growing diversity of heterogeneous data sources—ranging from microscopic images and physicochemical properties to simulation outputs and textual data—multimodal machine learning (MML) has become a key technology for fusing different types of information and supporting reliable multi-objective decision-making. However, despite increasing research interest, there is still no unified and conceptually structured review on the application of MML methods in data-driven material selection. This paper provides a thorough and structured overview of the field, organized using a novel multidimensional classification framework based on data types, integration levels, learning paradigms, and decision-making tasks. Guided by this framework, we critically evaluate the strengths, limitations, and general applicability of existing methods, and identify current trends and research gaps. Beyond qualitative synthesis, we quantify the cited corpus (N = 94) by publication year, method family, and application area, and collate an empirical fusion-evidence table that reports each study’s quantified gain together with its boundary conditions. We further analyze the main challenges and present a forward-looking research agenda covering self-supervised learning, knowledge-enhanced models, and interpretable human–AI collaboration. This review aims to offer a conceptual framework and a solid reference for building intelligent multimodal material selection systems.
A materials property hierarchy is introduced, from intrinsic, composition-determined properties to extrinsic, processing-dependent performance, to clarify deployment constraints and distinguish structural, physical and deployment novelty.
Xianyuan Liu, Charles Anjah, Benjamin E. Jolly et al.· Journal of Physics Materials· 0 citations
Automated machine learning (AutoML) reduces the cost of developing high-performing models by automating pipeline design, model selection, and hyperparameter optimization. As data sets increasingly combine heterogeneous sources (e.g., text, images, audio, and structured/tabular records), AutoML is extending to multimodal settings, where performance depends on representation learning and cross-modal fusion as well as model search. This survey reviews automated multimodal machine learning (MMAutoML), distinguishing core systems that automate both representation and fusion decisions from near-core systems, partial multimodal AutoML tools, and adjacent multimodal ML infrastructure. We organize multimodal learning around early, late, and hybrid fusion paradigms and discuss trade-offs in robustness, interpretability, and deployment. We then curate and compare representative open-source and commercial MMAutoML frameworks, summarize supported modalities, and characterize the ecosystem via time-evolution and similarity-based groupings. Finally, we overview applications in healthcare, autonomous systems, finance, and e-commerce, and highlight open challenges in modality alignment, missing or degraded inputs, efficiency and scalability, reproducibility, and governance (privacy, bias, and monitoring). We conclude with practical guidance and research directions toward reliable, end-to-end MMAutoML.
The intrinsic challenges of accessing historical dark data in materials science are described and contrasted with timely opportunities for leveraging massive amounts of experimental data from laboratories in going forwards; by exploiting electronic-lab notebooks, high-throughput experiments, and digital-twin technologies.
Jacqueline M. Cole· Advances in Materials· 0 citations
This work will present the current work on inverse material design, where AI methods—particularly generative pretrained transformers—are used to predict new material candidates based on desired properties, pushing the boundaries of materials innovation.
I. Gonzales, R. Ullberg, Andrew H Salij et al.· ECS Meeting Abstracts· 0 citations
This study examines the application of ML techniques to composite materials, particularly for predicting fracture toughness, characterizing damage, and optimizing mechanical properties and reveals significant relationships between fracture toughness and important input parameters.
Periyasamy Chitra· Building Materials and Engin...· 0 citations
Highlights A PE-ECC database comprises 383 material-level records from 90 literature sources. The MoE model predicts four PE-ECC properties with R2 values of 0.950–0.971. SHAP, ALE and response maps distinguish strength trends from tensile deformation. Binder, W/B, S/B and fiber variables show property-specific relationships. Support-filtered screening yields database-supported candidates for laboratory validation. Abstract Featuring considerable tensile ductility and multiple cracking behavior, polyethylene fiber-reinforced engineered cementitious composites (PE-ECCs) are promising cement-based materials for engineering construction. However, establishing accurate design models for evaluating the mechanical properties of PE-ECC is a challenging task owing to the complex material components. This study presents an interpretable data-driven framework for predicting the mechanical properties of PE-ECC using mixture-of-experts (MoE) learning. A database comprising 383 deduplicated material-level records from 90 verified literature sources was compiled for modeling the compressive strength, ultimate tensile strain, ultimate tensile strength and first-cracking tensile strength of PE-ECC. An MoE prediction model was developed by integrating XGBoost, LightGBM, CatBoost, WDBPANN and TabPFN through out-of-fold stacking and learned gating. The model achieved coefficient of determination (R2) values of 0.971, 0.950, 0.970 and 0.954 for the four mechanical properties, respectively. Shapley additive explanations (SHAP), accumulated local effects (ALE) and response maps were used to examine the fitted nonlinear associations between the reported mixture variables and each target property. Based on these relationships, support-filtered virtual screening was conducted within the database-supported design space to identify candidate mixtures for subsequent experimental verification. The framework links target-specific prediction with mixture-response interpretation and confines screening to regions supported by reported PE-ECC mixtures.