Artificial intelligence-driven approaches for materials design and discovery
The DOI matches its Crossref record; 99 of the 99 references it cites were found in a registry or an index.
Abstract
Materials design is an important component of modern science and technology, yet traditional approaches rely heavily on trial and error and can be inefficient. Computational techniques, enhanced by modern artificial intelligence, have reshaped the landscape of designing new materials. Among these approaches, inverse design has shown great promise in designing materials that meet specific property requirements. In this Review, we present key computational advances in materials design over the past few decades. We follow the evolution of relevant materials design techniques, from high-throughput forward machine learning methods and evolutionary algorithms, to advanced artificial intelligence strategies such as reinforcement learning and deep generative models. We highlight the paradigm shift from conventional screening approaches to inverse generation driven by deep generative models. Finally, we discuss current challenges and future perspectives of materials inverse design. This Review may serve as a brief guide to the approaches, progress and outlook of designing future functional materials with technological relevance.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“To the best of our knowledge, no prior review has isolated simulation-driven, toolorchestrating AI agent workflows for materials discovery as a distinct evidence base with study-level extraction of autonomy, validation, and reproducibility [3-14].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [4]
In the citing paper
“The central challenge is not only to predict a property, but to manage the sequence of reasoning and action that connects a scientific objective to executable computational or experimental steps [2-10].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [4]
In the citing paper
“Reviews of AI-driven materials design focus more broadly on generative models, inverse design, reinforcement learning, and materials-discovery acceleration [4].”
From From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery (2026) · cites this paper as [4]
Citations
Cites
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation
Jain, A., Ong, S. P., Hautier, G., Chen, W., Richards, W. D., Dacek, S., Cholia, S., Gunter, D., Skinner, D., Ceder, G., et al. (2013) · APL Materials · doi:10.1063/1.4812323 · Cited by 13,094 · cited 1×
“1a, the forward screening typically starts from collecting candidate materials, often sourced from open-source databases 17 .”
Taking the Human Out of the Loop: A Review of Bayesian Optimization
Shahriari, B., Swersky, K., Wang, Z., Adams, R. P., & de Freitas, N. (2016) · Proceedings of the IEEE · doi:10.1109/jproc.2015.2494218 · Cited by 6,049 · cited 1×
“BO is a powerful framework for global optimization in scenarios where data are scarce or expensive to acquire 45 .”
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602 · cited 2×
“For example, CrystalFormer is an autoregressive transformer that generates crystals by leveraging space group symmetries, achieving efficient data usage and property-guided materials design 94 , Similarly, Crystal Composition Transformer 95 , a self-supervised language model tailored for crystal compositions, is capab…”
Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules
Gómez-Bombarelli, R., Wei, J. N., Duvenaud, D., Hernández-Lobato, J. M., Sánchez-Lengeling, B., Sheberla, D., Aguilera-Iparraguirre, J., Hirzel, T. D., Adams, R. P., & Aspuru-Guzik, A. (2018) · ACS Central Science · doi:10.1021/acscentsci.7b00572 · Cited by 3,106 · cited 2×
“One of the early attempts develops VAE to generate novel molecules for drug and materials design 57 .”
CALYPSO: A method for crystal structure prediction
Wang, Y., Lv, J., Zhu, L., & Ma, Y. (2012) · Computer Physics Communications · doi:10.1016/j.cpc.2012.05.008 · Cited by 2,897 · cited 1×
“For example, PSO has been used as a crystal structure predictor, implemented in the CALYPSO package 38 .”
Xie, T., & Grossman, J. C. (2018) · Physical Review Letters · doi:10.1103/physrevlett.120.145301 · Cited by 2,828 · cited 3×
“This poses a challenge for forward screening to discover novel materials with properties beyond existing trends. b , Crystal graph convolutional neural networks as a prominent framework of a GNN for materials property prediction and screening 27 .”
InfluentialCrystal structure prediction via particle-swarm optimization
Wang, Y., Lv, J., Zhu, L., & Ma, Y. (2010) · Physical Review B · doi:10.1103/physrevb.82.094116 · Cited by 2,747 · cited 1×
“The CALYPSO code efficiently optimizes the free-energy surface and generates crystal structures based solely on chemical composition 39 .”
Crystal structure prediction using ab initio evolutionary techniques: Principles and applications
Oganov, A. R., & Glass, C. W. (2006) · The Journal of Chemical Physics · doi:10.1063/1.2210932 · Cited by 2,596 · cited 2×
“Early attempts such as genetic algorithms and Monte Carlo tree search (MCTS) laid the groundwork for inverse design development 8,9 .”
InfluentialAFLOW: An automatic framework for high-throughput materials discovery
Curtarolo, S., Setyawan, W., Hart, G. L. W., Jahnatek, M., Chepulskii, R. V., Taylor, R. H., Wang, S., Xue, J., Yang, K., Levy, O., et al. (2012) · Computational Materials Science · doi:10.1016/j.commatsci.2012.02.005 · Cited by 1,531 · cited 1×
“Automated frameworks such as Atomate 18 , JARVIS-Tools 19 and AFLOW 20 greatly facilitate such forward screening processes by streamlining data preparation, density functional theory (DFT) calculations and post-analysis.”
The evolution of multiferroics
Fiebig, M., Lottermoser, T., Meier, D., & Trassin, M. (2016) · Nature Reviews Materials · doi:10.1038/natrevmats.2016.46 · Cited by 1,469 · cited 1×
“…great promise for multi-objective optimization tasks, such as the simultaneous improvement of thermoelectric efficiency 120 or multiferroic coupling 121 , and balancing material functionalities with environmental sustainability 122 , which have been extremely challenging through traditional…”
Computational Design of Hierarchically Structured Materials
Olson, G. B. (1997) · Science · doi:10.1126/science.277.5330.1237 · Cited by 1,229 · cited 2×
“Around the turn of the millennium, computational techniques, such as hierarchically structured materials design 1 and integrated computational materials engineering 2 , transformed materials discovery by using computational models to automatically identify promising material candidates.”
Bayesian reaction optimization as a tool for chemical synthesis
Shields, B. J., Stevens, J., Li, J., Parasram, M., Damani, F., Alvarado, J. I. M., Janey, J. M., Adams, R. P., & Doyle, A. G. (2021) · Nature · doi:10.1038/s41586-021-03213-y · Cited by 1,063 · cited 1×
“A promising application of BO lies in reaction optimization 46 , shown in Fig.”
An autonomous laboratory for the accelerated synthesis of inorganic materials
Szymanski, N. J., Rendy, B., Fei, Y., Kumar, R. E., He, T., Milsted, D., McDermott, M. J., Gallant, M., Cubuk, E. D., Merchant, A., et al. (2023) · Nature · doi:10.1038/s41586-023-06734-w · Cited by 1,041 · cited 2×
“A-Lab represents an autonomous laboratory designed to accelerate the synthesis of novel inorganic materials by seamlessly integrating ML algorithms and advanced robotics.”
CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling
Deng, B., Zhong, P., Jun, K., Riebesell, J., Han, K., Bartel, C. J., & Ceder, G. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-023-00716-3 · Cited by 944 · cited 1×
“Recent progress has led to the emergence of pretrained universal MLIPs such as CHGNet 21 , MACE-MP-0 22 , MatterSim 23 and SevenNet-0 24 , which”
Small data machine learning in materials science
Xu, P., Ji, X., Li, M., & Lu, W. (2023) · npj Computational Materials · doi:10.1038/s41524-023-01000-z · Cited by 866 · cited 1×
“Many materials domains, such as unconventional superconductors and complex oxides, suffer from small or imbalanced datasets, making it difficult to train reliable and generalizable models 129 .”
Self-citationGraph neural networks for materials science and chemistry
Reiser, P., Neubert, M., Eberhard, A., Torresi, L., Zhou, C., Shao, C., Metni, H., van Hoesel, C., Schopmans, H., Sommer, T., et al. (2022) · Communications Materials · doi:10.1038/s43246-022-00315-6 · Cited by 860 · cited 1×
“Numerous structural representations are proposed for materials and summarized in several reviews 25,26 .”
High-performance thermoelectrics and challenges for practical devices
Yan, Q., & Kanatzidis, M. G. (2022) · Nature Materials · doi:10.1038/s41563-021-01109-w · Cited by 846 · cited 1×
“Generative models show great promise for multi-objective optimization tasks, such as the simultaneous improvement of thermoelectric efficiency 120 or multiferroic coupling 121 , and balancing material functionalities with environmental sustainability 122 , which have been extremely challenging…”
Structured information extraction from scientific text with large language models
Dagdelen, J., Dunn, A., Lee, S., Walker, N., Rosen, A. S., Ceder, G., Persson, K. A., & Jain, A. (2024) · Nature Communications · doi:10.1038/s41467-024-45563-x · Cited by 771 · cited 1×
“In addition, recent advances in LLMs open new opportunities for automating literature mining and curating experimental databases at scale 132 .”
Atomistic Line Graph Neural Network for improved materials property predictions
Choudhary, K., & DeCost, B. (2021) · npj Computational Materials · doi:10.1038/s41524-021-00650-1 · Cited by 758 · cited 1×
“Combining a diffusion model, a formation energy prediction network and the atomistic line graph neural network (ALIGNN) framework 117 for superconductivity evaluation, together with pretraining and fine-tuning strategies, and Inverse design has also been successfully applied to MOFs, leading to…”
Crystal structure prediction from first principles
Woodley, S. M., & Catlow, R. (2008) · Nature Materials · doi:10.1038/nmat2321 · Cited by 733 · cited 1×
“Early attempts such as genetic algorithms and Monte Carlo tree search (MCTS) laid the groundwork for inverse design development 8,9 .”
Structure prediction drives materials discovery
Oganov, A. R., Pickard, C. J., Zhu, Q., & Needs, R. J. (2019) · Nature Reviews Materials · doi:10.1038/s41578-019-0101-8 · Cited by 717
A foundation model for atomistic materials chemistry
Batatia, I., Benner, P., Chiang, Y., Elena, A. M., Kovács, D. P., Riebesell, J., Advincula, X. R., Asta, M., Avaylon, M., Baldwin, W. J., et al. (2024) · arXiv · doi:10.48550/arxiv.2401.00096 · Cited by 604 · cited 4×
“https://doi.org/10.1038/s41563-025-02403-7 under specific conditions, such as energy above the convex hull and phonon dispersion computed by MLIPs 22 .”
InfluentialKnaster, J., Moeslang, A., & Muroga, T. (2016) · Nature Physics · doi:10.1038/nphys3735 · Cited by 553 · cited 1×
“Current generative models primarily focus on interpolation within known datasets, limiting their ability to uncover groundbreaking materials such as room-temperature superconductors 123 , high zT thermoelectrics 124 and nuclear fusion materials 125 .”
Choudhary, K., Garrity, K. F., Reid, A. C. E., DeCost, B., Biacchi, A. J., Hight Walker, A. R., Trautt, Z., Hattrick-Simpers, J., Kusne, A. G., Centrone, A., et al. (2020) · npj Computational Materials · doi:10.1038/s41524-020-00440-1 · Cited by 541 · cited 1×
“Automated frameworks such as Atomate 18 , JARVIS-Tools 19 and AFLOW 20 greatly facilitate such forward screening processes by streamlining data preparation, density functional theory (DFT) calculations and post-analysis.”
Physics-Inspired Structural Representations for Molecules and Materials
Musil, F., Grisafi, A., Bartók, A. P., Ortner, C., Csányi, G., & Ceriotti, M. (2021) · Chemical Reviews · doi:10.1021/acs.chemrev.1c00021 · Cited by 523 · cited 1×
“Numerous structural representations are proposed for materials and summarized in several reviews 25,26 .”
A generative model for inorganic materials design
Zeni, C., Pinsler, R., Zügner, D., Fowler, A., Horton, M., Fu, X., Wang, Z., Shysheya, A., Crabbé, J., Ueda, S., et al. (2025) · Nature · doi:10.1038/s41586-025-08628-5 · Cited by 518 · cited 7×
“Furthermore, MatterGen can be fine-tuned with a small amount of labelled data to achieve conditional generation on various constraints, including chemical elements, symmetry and material properties such as magnetic density and bulk modulus 77 (Fig.”
InfluentialInverse design in search of materials with target functionalities
Zunger, A. (2018) · Nature Reviews Chemistry · doi:10.1038/s41570-018-0121 · Cited by 513
Review of current high-ZT thermoelectric materials
Wei, J., Yang, L., Ma, Z., Song, P., Zhang, M., Ma, J., Yang, F., & Wang, X. (2020) · Journal of Materials Science · doi:10.1007/s10853-020-04949-0 · Cited by 413 · cited 1×
“Current generative models primarily focus on interpolation within known datasets, limiting their ability to uncover groundbreaking materials such as room-temperature superconductors 123 , high zT thermoelectrics 124 and nuclear fusion materials 125 .”
Inverse design of nanoporous crystalline reticular materials with deep generative models
Yao, Z., Sánchez-Lengeling, B., Bobbitt, N. S., Bucior, B. J., Kumar, S. G. H., Collins, S. P., Burns, T., Woo, T. K., Farha, O. K., Snurr, R. Q., et al. (2021) · Nature Machine Intelligence · doi:10.1038/s42256-020-00271-1 · Cited by 400 · cited 1×
“Beyond these cases, VAEs have also been applied to alloy microstructure prediction 61 , high-entropy alloys 62 and nanoporous crystalline materials 63 .”
Mathew, K., Montoya, J. H., Faghaninia, A., Dwarakanath, S., Aykol, M., Tang, H., Chu, I.-H., Smidt, T., Bocklund, B., Horton, M., et al. (2017) · Computational Materials Science · doi:10.1016/j.commatsci.2017.07.030 · Cited by 358 · cited 1×
“Automated frameworks such as Atomate 18 , JARVIS-Tools 19 and AFLOW 20 greatly facilitate such forward screening processes by streamlining data preparation, density functional theory (DFT) calculations and post-analysis.”
Designing a New Material World
Olson, G. B. (2000) · Science · doi:10.1126/science.288.5468.993 · Cited by 344 · cited 1×
“Around the turn of the millennium, computational techniques, such as hierarchically structured materials design 1 and integrated computational materials engineering 2 , transformed materials discovery by using computational models to automatically identify promising material candidates.”
Artificial Chemist: An Autonomous Quantum Dot Synthesis Bot
Epps, R. W., Bowen, M. S., Volk, A. A., Abdel‐Latif, K., Han, S., Reyes, K. G., Amassian, A., & Abolhasani, M. (2020) · Advanced Materials · doi:10.1002/adma.202001626 · Cited by 323 · cited 1×
“Other notable efforts combining experimental techniques with inverse design include, but are not limited to, quantum dot synthesis 55 and AlphaFlow, an RL-based platform optimized for synthetic pathways in the shell growth of core–shell semiconductor nanoparticles 56 .”
Generative models for molecular discovery: Recent advances and challenges
Bilodeau, C., Jin, W., Jaakkola, T., Barzilay, R., & Jensen, K. F. (2022) · WIREs Computational Molecular Science · doi:10.1002/wcms.1608 · Cited by 318
Genetic algorithms for computational materials discovery accelerated by machine learning
Jennings, P. C., Lysgaard, S., Hummelshøj, J. S., Vegge, T., & Bligaard, T. (2019) · npj Computational Materials · doi:10.1038/s41524-019-0181-4 · Cited by 269 · cited 1×
“This approach reduced the number of DFT calculations by up to 50-fold compared with traditional genetic algo-rithm approaches and was used to identify optimal chemical ordering within binary alloy nanoparticles 37 .”
Generative Adversarial Networks for Crystal Structure Prediction
Kim, S., Noh, J., Gu, G. H., Aspuru-Guzik, A., & Jung, Y. (2020) · ACS Central Science · doi:10.1021/acscentsci.0c00426 · Cited by 265 · cited 1×
“O ternary systems, achieving notable success in discovering new compositions and predicting their stability through high-throughput virtual screening 67 .”
Inverse design of nonlinear mechanical metamaterials via video denoising diffusion models
Bastek, J.-H., & Kochmann, D. M. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-023-00762-x · Cited by 252 · cited 1×
“Instead, it predicts the expected deformation path and internal stress distributions, achieving remarkable agreement with finite element simulations 70 .”
Kang, Y., & Kim, J. (2024) · Nature Communications · doi:10.1038/s41467-024-48998-4 · Cited by 236 · cited 2×
“Similar work includes ChatMOF 97 , which is built to predict and generate metal–organic frameworks (MOFs); Atom-GPT 98 , a transformer-based architecture to predict material properties and generate atomic structure combined with chemical and structural text descriptions; and CrystaLLM 99 , a…”
Park, Y., Kim, J., Hwang, S., & Han, S. (2024) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.4c00190 · Cited by 228
First-principles structural design of superhard materials
Zhang, X., Wang, Y., Lv, J., Zhu, C., Li, Q., Zhang, M., Li, Q., & Ma, Y. (2013) · The Journal of Chemical Physics · doi:10.1063/1.4794424 · Cited by 225 · cited 1×
“N systems, PSO reproduced both experimentally and theoretically known structures 40 .”
Volk, A. A., Epps, R. W., Yonemoto, D. T., Masters, B. S., Castellano, F. N., Reyes, K. G., & Abolhasani, M. (2023) · Nature Communications · doi:10.1038/s41467-023-37139-y · Cited by 215 · cited 1×
“Other notable efforts combining experimental techniques with inverse design include, but are not limited to, quantum dot synthesis 55 and AlphaFlow, an RL-based platform optimized for synthetic pathways in the shell growth of core–shell semiconductor nanoparticles 56 .”
Discovery and Optimization of Materials Using Evolutionary Approaches
Le, T. C., & Winkler, D. A. (2016) · Chemical Reviews · doi:10.1021/acs.chemrev.5b00691 · Cited by 204 · cited 1×
“Evolutionary algorithms evaluate the parameter sets through a fitness function of the metrics quantifying the performance with respect to the specific design objectives, including catalytic performance, hardness, synthesizability and magnetization 33,34 .”
Solar fuels photoanode materials discovery by integrating high-throughput theory and experiment
Yan, Q., Yu, J., Suram, S. K., Zhou, L., Shinde, A., Newhouse, P. F., Chen, W., Li, G., Persson, K. A., Gregoire, J. M., et al. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1619940114 · Cited by 192 · cited 1×
“Similar high-throughput experiments have also been applied in solar fuels photoanode materials discovery 111 and multi-property high-entropy alloys 112 .”
Ren, Z., Tian, S. I. P., Noh, J., Oviedo, F., Xing, G., Li, J., Liang, Q., Zhu, R., Aberle, A. G., Sun, S., et al. (2022) · Matter · doi:10.1016/j.matt.2021.11.032 · Cited by 188 · cited 1×
“In thermoelectrics, two inverse-designed crystals with state-of-the-art power factors were reported 59 , comparable to the best known thermoelectric materials.”
Bayesian optimization with adaptive surrogate models for automated experimental design
Lei, B., Kirk, T., Bhattacharya, A., Pati, D., Qian, X., Arróyave, R., & Mallick, B. K. (2021) · npj Computational Materials · doi:10.1038/s41524-021-00662-x · Cited by 183 · cited 1×
“BO is increasingly used as an adaptive surrogate model for automated experimental design 47 .”
Inverse design workflow discovers hole-transport materials tailored for perovskite solar cells
Wu, J., Torresi, L., Hu, M., Reiser, P., Zhang, J., Rocha-Ortiz, J. S., Wang, L., Xie, Z., Zhang, K., Park, B.-W., et al. (2024) · Science · doi:10.1126/science.ads0901 · Cited by 183 · cited 1×
“Recently, a closed-loop inverse design strategy has been demonstrated 110 , which integrates high-throughput synthesis and ML to discover high-performance organic hole-transport materials for perovskite solar cells.”
Data-driven discovery of 2D materials by deep generative models
Lyngby, P., & Thygesen, K. S. (2022) · npj Computational Materials · doi:10.1038/s41524-022-00923-3 · Cited by 157 · cited 1×
“Applications of CDVAE to 2D materials have demonstrated its capacity to generate novel stable structures with high chemical and structural diversity 73 .”
Crystal structure generation with autoregressive large language modeling
Antunes, L. M., Butler, K. T., & Grau-Crespo, R. (2024) · Nature Communications · doi:10.1038/s41467-024-54639-7 · Cited by 157 · cited 2×
“…98 , a transformer-based architecture to predict material properties and generate atomic structure combined with chemical and structural text descriptions; and CrystaLLM 99 , a methodology for the versatile generation of crystal structures, based on the autoregressive LLM of the CIF format.”
Song, Y., Dhariwal, P., Chen, M., & Sutskever, I. (2023) · arXiv · doi:10.48550/arxiv.2303.01469 · Cited by 155 · cited 1×
“We also briefly discuss flow matching 86 and consistency models 87 , which are emerging generative models beyond diffusion models.”
Long, T., Fortunato, N. M., Opahle, I., Zhang, Y., Samathrakis, I., Shen, C., Gutfleisch, O., & Zhang, H. (2021) · npj Computational Materials · doi:10.1038/s41524-021-00526-4 · Cited by 152 · cited 1×
“Another deep convolutional GAN links crystal images with structural and formation energy constraints, enabling formation energy-guided structure generation 68 .”
Combinatorial synthesis for AI-driven materials discovery
Gregoire, J. M., Zhou, L., & Haber, J. A. (2023) · Nature Synthesis · doi:10.1038/s44160-023-00251-4 · Cited by 146 · cited 1×
“16.”
Applications of natural language processing and large language models in materials discovery
Jiang, X., Wang, W., Tian, S., Wang, H., Lookman, T., & Su, Y. (2025) · npj Computational Materials · doi:10.1038/s41524-025-01554-0 · Cited by 145 · cited 1×
“The advances in natural language processing techniques and the development of LLMs are improving the efficient extraction and utilization of information 93 .”
High-temperature superconductivity
Zhou, X., Lee, W.-S., Imada, M., Trivedi, N., Phillips, P., Kee, H., Törmä, P., & Eremets, M. I. (2021) · Nature Reviews Physics · doi:10.1038/s42254-021-00324-3 · Cited by 143 · cited 1×
“Current generative models primarily focus on interpolation within known datasets, limiting their ability to uncover groundbreaking materials such as room-temperature superconductors 123 , high zT thermoelectrics 124 and nuclear fusion materials 125 .”
Machine learning-based screening of complex molecules for polymer solar cells
Jørgensen, P. B., Mesta, M., Shil, S., García Lastra, J. M., Jacobsen, K. W., Thygesen, K. S., & Schmidt, M. N. (2018) · The Journal of Chemical Physics · doi:10.1063/1.5023563 · Cited by 135
Machine‐Learning Microstructure for Inverse Material Design
Pei, Z., Rozman, K. A., Doğan, Ö. N., Wen, Y., Gao, N., Holm, E. A., Hawk, J. A., Alman, D. E., & Gao, M. C. (2021) · Advanced Science · doi:10.1002/advs.202101207 · Cited by 135 · cited 1×
“Beyond these cases, VAEs have also been applied to alloy microstructure prediction 61 , high-entropy alloys 62 and nanoporous crystalline materials 63 .”
High-throughput discovery of high Curie point two-dimensional ferromagnetic materials
Kabiraj, A., Kumar, M., & Mahapatra, S. (2020) · npj Computational Materials · doi:10.1038/s41524-020-0300-2 · Cited by 126 · cited 2×
“ML speeds up predictions while balancing computational cost (red arrows) and accuracy (green arrows). d , The screening workflow for identifying 2D ferromagnetic (FM) materials 32 .”
AI-Driven Inverse Design of Materials: Past, Present, and Future
Han, X., Wang, X.-D., 徐, M.-Y. 孟. X., Zhen, F., 姚, B.-W. 博. Y., Guo, P.-J., Gao, Z.-F., & Lu, Z.-Y. (2025) · Chinese Physics Letters · doi:10.1088/0256-307x/42/2/027403 · Cited by 123
Space group informed transformer for crystalline materials generation
Cao, Z., Luo, X., Lv, J., & Wang, L. (2025) · 中国科学通报:英文版 · doi:10.1016/j.scib.2025.09.035 · Cited by 111 · cited 1×
“…transformer that generates crystals by leveraging space group symmetries, achieving efficient data usage and property-guided materials design 94 , Similarly, Crystal Composition Transformer 95 , a self-supervised language model tailored for crystal compositions, is capable of unsupervised…”
High‐throughput computational materials screening and discovery of optoelectronic semiconductors
Luo, S., Li, T., Wang, X., Faizan, M., & Zhang, L. (2021) · WIREs Computational Molecular Science · doi:10.1002/wcms.1489 · Cited by 100 · cited 2×
“It has also been used to discover new optoelectronic semiconductors, including light absorbers, transparent conductors, light-emitting diode materials and photovoltaic materials, by screening for electronic properties 29 (Fig.”
Exploring High Thermal Conductivity Amorphous Polymers Using Reinforcement Learning
Ma, R., Zhang, H., & Luo, T. (2022) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.1c23610 · Cited by 87 · cited 1×
“Recent RL applications in inverse materials design include the compositional design of multicomponent alloys 50 , high thermal conductivity amorphous polymers 51 , and the design of mechanical metamaterials with nonlinear deformation responses 52 .”
WyCryst: Wyckoff inorganic crystal generator framework
Zhu, R., Nong, W., Yamazaki, S., & Hippalgaonkar, K. (2024) · Matter · doi:10.1016/j.matt.2024.05.042 · Cited by 87 · cited 1×
“VAEs have also been modified with symmetric constraints for material generation, such as WyCryst with Wyckoff representation 65 and SyMat with invariance of symmetric operation 66 .”
Buehler, M. J. (2024) · ACS Engineering Au · doi:10.1021/acsengineeringau.3c00058 · Cited by 82
AtomGPT: Atomistic Generative Pretrained Transformer for Forward and Inverse Materials Design
Choudhary, K. (2024) · The Journal of Physical Chemistry Letters · doi:10.1021/acs.jpclett.4c01126 · Cited by 72 · cited 2×
“Similar work includes ChatMOF 97 , which is built to predict and generate metal–organic frameworks (MOFs); Atom-GPT 98 , a transformer-based architecture to predict material properties and generate atomic structure combined with chemical and structural text descriptions; and CrystaLLM 99 , a…”
Brown, N., Garland, A., Fadel, G., & Li, G. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106998 · Cited by 71 · cited 1×
“Recent RL applications in inverse materials design include the compositional design of multicomponent alloys 50 , high thermal conductivity amorphous polymers 51 , and the design of mechanical metamaterials with nonlinear deformation responses 52 .”
Inverse design of porous materials: a diffusion model approach
Park, J., Gill, A. P. S., Moosavi, S. M., & Kim, J. (2024) · Journal of Materials Chemistry A · doi:10.1039/d3ta06274k · Cited by 71 · cited 1×
“Furthermore, conditional generation within latent spaces has also been used for porous materials, where interpolation of latent vectors has uncovered novel zeolite structures optimized for properties such as void fractions and adsorption capacity 75 .”
Bayesian optimization with experimental failure for high-throughput materials growth
Wakabayashi, Y. K., Otsuka, T., Krockenberger, Y., Sawada, H., Taniyasu, Y., & Yamamoto, H. (2022) · npj Computational Materials · doi:10.1038/s41524-022-00859-8 · Cited by 70 · cited 1×
“Recent developments include multi-scale batch BO for efficient navigation of microstructure spaces under optimized processing conditions 48 and a variant capable of handling missing data when target materials fail to form owing to suboptimal growth parameters 49 .”
MDTS: automatic complex materials design using Monte Carlo tree search
M. Dieb, T., Ju, S., Yoshizoe, K., Hou, Z., Shiomi, J., & Tsuda, K. (2017) · Science and Technology of Advanced Materials · doi:10.1080/14686996.2017.1344083 · Cited by 67 · cited 1×
“MCTS further contributes to solving complex material design problems such as alloy systems 44 .”
Con-CDVAE: A method for the conditional generation of crystal structures
Ye, C., Weng, H., & Wu, Q. (2024) · Computational Materials Today · doi:10.1016/j.commt.2024.100003 · Cited by 66 · cited 1×
“A extension of CDVAE, termed Con-CDVAE, realized the conditional generation of materials structures by carefully encoding target properties such as formation energy and bandgap, guiding the latent-based model to generate feasible structures with the target constraints 74 .”
High-throughput design of magnetic materials
Zhang, H. (2021) · Electronic Structure · doi:10.1088/2516-1075/abbb25 · Cited by 60 · cited 1×
“In addition, forward screening has facilitated magnetic materials design for applications such as permanent magnets, spintronics and magnetic refrigeration 31 .”
Identification of post-pyrite phase transitions in SiO 2 by a genetic algorithm
Wu, S., Umemoto, K., Ji, M., Wang, C. Z., Ho, K. M., & Wentzcovitch, R. M. M. (2011) · Physical Review B · doi:10.1103/physrevb.83.184102 · Cited by 60 · cited 1×
“For example, genetic algorithms were combined with DFT to predict an Fe 2 P phase as the first post-pyrite phase of SiO 2 at low temperatures 35 .”
Transforming the synthesis of carbon nanotubes with machine learning models and automation
Li, Y., Wang, S., Lv, Z., Wang, Z., Zhao, Y., Xie, Y., Xu, Y., Qian, L., Yang, Y., Zhao, Z., et al. (2024) · Matter · doi:10.1016/j.matt.2024.11.007 · Cited by 58 · cited 1×
“Specifically, the AI-recommended TiPt catalyst outperforms traditional Fe catalysts, achieving higher-accuracy density control in the synthesis of horizontally aligned carbon nanotube arrays 104 .”
Compositional design of multicomponent alloys using reinforcement learning
Xian, Y., Dang, P., Tian, Y., Jiang, X. F., Zhou, Y., Ding, X., Sun, J., Lookman, T., & Xue, D. (2024) · Acta Materialia · doi:10.1016/j.actamat.2024.120017 · Cited by 54 · cited 1×
“Recent RL applications in inverse materials design include the compositional design of multicomponent alloys 50 , high thermal conductivity amorphous polymers 51 , and the design of mechanical metamaterials with nonlinear deformation responses 52 .”
Wei, L., Li, Q., Song, Y., Stefanov, S., Dong, R., Fu, N., Siriwardane, E. M. D., Chen, F., & Hu, J. (2024) · Advanced Science · doi:10.1002/advs.202304305 · Cited by 49 · cited 2×
“For example, CrystalFormer is an autoregressive transformer that generates crystals by leveraging space group symmetries, achieving efficient data usage and property-guided materials design 94 , Similarly, Crystal Composition Transformer 95 , a self-supervised language model tailored for crystal compositions, is capab…”
Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models
Kim, S.-M., Schrier, J., & Jung, Y. (2025) · Angewandte Chemie International Edition · doi:10.1002/anie.202423950 · Cited by 47 · cited 1×
“For example, fine-tuned LLMs have been developed for synthesizability prediction and interpretable rule extraction from human-readable text descriptions of crystal structures 100 , offering explainable reasoning for structure feasibility.”
Honarmandi, P., Attari, V., & Arróyave, R. (2022) · Computational Materials Science · doi:10.1016/j.commatsci.2022.111417 · Cited by 47 · cited 1×
“Recent developments include multi-scale batch BO for efficient navigation of microstructure spaces under optimized processing conditions 48 and a variant capable of handling missing data when target materials fail to form owing to suboptimal growth parameters 49 .”
Virtual node graph neural network for full phonon prediction
Okabe, R., Chotrattanapituk, A., Boonkird, A., Andrejevic, N., Fu, X., Jaakkola, T., Song, Q., Nguyen, T., Drucker, N. C., Mu, S., et al. (2024) · Nature Computational Science · doi:10.1038/s43588-024-00661-0 · Cited by 47 · cited 1×
“1b), although it is worth noting that graph nodes have a broader meaning beyond atomic representations 28 .”
Self-citationA deep generative modeling architecture for designing lattice-constrained perovskite materials
Chenebuah, E. T., Nganbe, M., & Tchagang, A. (2024) · npj Computational Materials · doi:10.1038/s41524-024-01381-9 · Cited by 45 · cited 1×
“…letting the generator propose a crystal structure directly from a fixed latent vector, this approach allows the generator to refine nearby latent vectors under lattice constraints, which addresses the low-quality reconstruction problem of VAEs that can lead to materials with lower symmetries 69 .”
Structural constraint integration in a generative model for the discovery of quantum materials
Okabe, R., Cheng, M., Chotrattanapituk, A., Mandal, M., Mak, K., Carrizales, D. C., Hung, N. T., Fu, X., Han, B., Wang, Y., et al. (2025) · Nature Materials · doi:10.1038/s41563-025-02355-y · Cited by 42
Self-citationBeyond designer's knowledge: Generating materials design hypotheses via a large language model
Liu, Q., Polak, M. P., Kim, S. Y., Shuvo, M. A. A., Deodhar, H. S., Han, J., Morgan, D. D., & Oh, H. (2025) · Acta Materialia · doi:10.1016/j.actamat.2025.121307 · Cited by 40
Coevolutionary search for optimal materials in the space of all possible compounds
Allahyari, Z., & Oganov, A. R. (2020) · npj Computational Materials · doi:10.1038/s41524-020-0322-9 · Cited by 38 · cited 1×
“Evolutionary algorithms evaluate the parameter sets through a fitness function of the metrics quantifying the performance with respect to the specific design objectives, including catalytic performance, hardness, synthesizability and magnetization 33,34 .”
InvDesFlow-AL: active learning-based workflow for inverse design of functional materials
Han, X.-Q., Guo, P., Gao, Z.-F., Sun, H., & Lu, Z.-Y. (2025) · npj Computational Materials · doi:10.1038/s41524-025-01830-z · Cited by 38 · cited 1×
“Recently, studies on materials inverse design for discovering high-T c superconductors were reported 115,116 , including the model InvDesFlow and its extension InvDesFlow-AL with active learning.”
Experimentally validated inverse design of multi-property Fe-Co-Ni alloys
Padhy, S. P., Chaudhary, V., Lim, Y.-F., Zhu, R., Thway, M., Hippalgaonkar, K., & Ramanujan, R. V. (2024) · iScience · doi:10.1016/j.isci.2024.109723 · Cited by 34 · cited 1×
“Similar high-throughput experiments have also been applied in solar fuels photoanode materials discovery 111 and multi-property high-entropy alloys 112 .”
Structural optimization of Pt–Pd–Au trimetallic nanoparticles by discrete particle swarm algorithms
Fan, T., Liu, T.-D., Zheng, J.-W., Shao, G., & Wen, Y. (2015) · Journal of Materials Science · doi:10.1007/s10853-015-8880-9 · Cited by 33 · cited 15×
“Evolutionary algorithms can be categorized into methods such as genetic algorithms, particle swarm optimization (PSO) and MCTS, each with a unique approach to solving nonlinear, high-dimensional and multimodal optimization problems.”
InfluentialA Continuous Action Space Tree search for INverse desiGn (CASTING) framework for materials discovery
Banik, S., Loeffler, T. D., Manna, S., Chan, H., Srinivasan, S., Darancet, P., Hexemer, A., & Sankaranarayanan, S. K. R. S. (2023) · npj Computational Materials · doi:10.1038/s41524-023-01128-y · Cited by 33 · cited 2×
“Another framework called continuous action space tree search for inverse design (CASTING) has been proposed for materials inverse design, using MCTS for continuous action space to navigate a complex energy landscape 43 (Fig.”
Han, X.-Q., Ouyang, Z., Guo, P., Sun, H., Gao, Z., & Lu, Z.-Y. (2025) · Chinese Physics Letters · doi:10.1088/0256-307x/42/4/047301 · Cited by 33 · cited 1×
“Recently, studies on materials inverse design for discovering high-T c superconductors were reported 115,116 , including the model InvDesFlow and its extension InvDesFlow-AL with active learning.”
SciAssess: Benchmarking LLM Proficiency in Scientific Literature Analysis
Cai, H., Cai, X., Chang, J., Li, S., Yao, L., Changxin, W., Gao, Z., Wang, H., Li, Y., Lin, M., et al. (2025) · Findings of the Association for Computational Linguistics: NAACL · doi:10.18653/v1/2025.findings-naacl.125 · Cited by 31
Inverse mapping of quantum properties to structures for chemical space of small organic molecules
Fallani, A., Sandonas, L. M., & Tkatchenko, A. (2024) · Nature Communications · doi:10.1038/s41467-024-50401-1 · Cited by 30 · cited 2×
“…by another small ML model to create a property map, in which the search and optimization algorithm can provide candidate latent vectors, from which the decoder can reconstruct them to obtain structures with the desired properties 57 . c , Jointly trained VAE between structure–property learning 64”
Tom, R., Gao, S., Yang, Y., Zhao, K., Bier, I., Buchanan, E. A., Zaykov, A., Havlas, Z., Michl, J., & Marom, N. (2023) · Chemistry of Materials · doi:10.1021/acs.chemmater.2c03444 · Cited by 29 · cited 2×
“…and MCTS, where a decision tree is expanded (green solid arrow) iteratively using simulations (blue dashed circle) and back-propagation (brown dashed arrow) to identify optimal candidates ( c ). d , Evolutionary routes that explore the candidate structures of tetracene in a genetic algorithm 36 .”
Generative Artificial Intelligence for Advancing Discovery and Design in Biomateriomics
Pugliese, R., Badini, S., Frontoni, E., & Regondi, S. (2025) · Intelligent Computing · doi:10.34133/icomputing.0117 · Cited by 28 · cited 1×
“This approach holds immense Review article https://doi.org/10.1038/s41563-025-02403-7 generalize across diverse material systems, enabling broader applicability in high-throughput workflows.”
InfluentialZhong, C., Zhang, J., Lu, X., Zhang, K., Liu, J., Hu, K., Chen, J., & Lin, X. (2023) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.3c00593 · Cited by 23
Rosafalco, L., Ponti, J. M. D., Iorio, L., Craster, R. V., Ardito, R., & Corigliano, A. (2023) · Scientific Reports · doi:10.1038/s41598-023-48927-3 · Cited by 22 · cited 1×
“For instance, RL has been successfully applied to optimize the design of graded metamaterials 53 for energy harvesting under realistic conditions, such as magnetic loading and random excitations (Fig.”
Long, Y., Zhong, C., Ma, X., Zhang, J., Yao, H., Liu, J., Hu, K., Zhang, Q., & Lin, X. (2025) · ACS Applied Materials & Interfaces · doi:10.1021/acsami.4c19494 · Cited by 19 · cited 1×
“High-performance thermoelectric materials have also been successfully synthesized with computational inverse design 113 .”
MagGen: A Graph-Aided Deep Generative Model for Inverse Design of Permanent Magnets
Mal, S., Seal, G., & Sen, P. (2024) · The Journal of Physical Chemistry Letters · doi:10.1021/acs.jpclett.4c00068 · Cited by 15 · cited 1×
“In magnetic materials design, 46 stable magnetic compounds were discovered through DFT validation 58 ; notably, 20 of these were within 0.25 eV per atom of the convex hull, indicating potential experimental synthesizability.”
InfluentialSearching the stable segregation configuration at the grain boundary by a Monte Carlo tree search
Kiyohara, S., & Mizoguchi, T. (2018) · The Journal of Chemical Physics · doi:10.1063/1.5023139 · Cited by 14 · cited 1×
“MCTS has been applied to identify stable structures, such as the segregation of silver impurities in copper, by searching only 1% of all the possible configurations 42 .”
Zeng, C., Khan, Z., & Post, N. (2024) · AI & Materials · doi:10.55092/aimat20250002 · Cited by 12 · cited 1×
“Beyond these cases, VAEs have also been applied to alloy microstructure prediction 61 , high-entropy alloys 62 and nanoporous crystalline materials 63 .”
Discovery of 2D Materials via Symmetry-Constrained Diffusion Model
Xu, S., Chu, S., Mrad, R., Zhang, Z., Li, Z., Jiao, R., & Chen, Y. (2025) · The Journal of Physical Chemistry C · doi:10.1021/acs.jpcc.4c08681 · Cited by 6 · cited 1×
“We note that a similar technique has also been used to generate 2D crystals with symmetry constraints 85 .”
How to bridge the gap between theory and experiments in biological physics
Trepat, X., & Alert, R. (2023) · Nature Physics · doi:10.1038/s41567-023-02276-0 · Cited by 5 · cited 1×
“A further complication lies in the gap between first-principles calculations and experimental observations, because models trained solely on high-quality first-principles data do not necessarily generalize well to real-world scenarios 130,131 .”
Are quantum materials economically and environmentally sustainable?
Boonkird, A., Cheng, M., Chotrattanapituk, A., Carrizales, D. C., Okabe, R., Drucker, N. C., Mandal, M., Nguyen, T., Yeo, J., Belosevich, V., et al. (2025) · Materials Today · doi:10.1016/j.mattod.2025.09.014 · Cited by 4 · cited 1×
“…promise for multi-objective optimization tasks, such as the simultaneous improvement of thermoelectric efficiency 120 or multiferroic coupling 121 , and balancing material functionalities with environmental sustainability 122 , which have been extremely challenging through traditional approaches.”
Self-citationMind the Gap Between Theory and Experiment
Kiselev, A., Kim, J., & Martin, O. J. F. (2023) · Springer eBooks · doi:10.1007/978-3-031-24617-3_21 · Cited by 3
Author response for "Establishing baselines for generative discovery of inorganic crystals"
Szymanski, N. J., & Bartel, C. J. (2025) · Publons reviews and discussion · doi:10.1039/d5mh00010f/v2/response1 · Cited by 1
Research works printed but not linked: 32
- Mnih, V. Playing Atari with deep reinforcement learning. Preprint at https://arxiv.org/abs/1312.5602 (2013). · arXiv
- Christiano, P. F. et al. Deep reinforcement learning from human preferences. Adv. Neural Inf. Process. Syst. 30, (2017).
- Kingma, D. P. Auto-encoding variational Bayes. Preprint at https://arxiv.org/abs/1312.6114 (2013). · arXiv
- Goodfellow, I. et al. Generative adversarial nets. Adv. Neural Inf. Process. Syst. 27, (2014).
- Van Den Oord, A., Kalchbrenner, N. & Kavukcuoglu, K. Pixel recurrent neural networks. In International Conference on Machine Learning 1747–1756 (PMLR, 2016).
- Ho, J., Jain, A. & Abbeel, P. Denoising diffusion probabilistic models. Adv. Neural Inf. Process. Syst. 33, 6840–6851 (2020).
- Yang, H. et al. MatterSim: a deep learning atomistic model across elements, temperatures and pressures. Preprint at https://arxiv.org/abs/2405.04967 (2024). · arXiv
- Carrete, J., Li, W., Mingo, N., Wang, S. & Curtarolo, S. Finding unprecedentedly low-thermal-conductivity half-Heusler semiconductors via high-throughput materials modeling. Phys. Rev. X 4, 011019 (2014).
- Luo, Y., Liu, C. & Ji, S. Towards symmetry-aware generation of periodic materials. Adv. Neural Inf. Process. Syst. 36, (2024).
- Fernandez-Zelaia, P. et al. Denoising diffusion probabilistic models for generative alloy design. Addit. Manuf. 94, 104478 (2024).
- Xie, T., Fu, X., Ganea, O.-E., Barzilay, R. & Jaakkola, T. Crystal diffusion variational autoencoder for periodic material generation. Preprint at https://arxiv.org/abs/2110.06197 (2021). · arXiv
- Jiao, R. et al. Crystal structure prediction by joint equivariant diffusion. Adv. Neural Inf. Process. Syst. 36, 17464–17497 (2023).
- Tangsongcharoen, K. et al. CrystalGRW: generative modeling of crystal structures with targeted properties via geodesic random walks. Preprint at https://arxiv.org/abs/2501.08998 (2025). · arXiv
- Yang, S. et al. Generative hierarchical materials search. Adv. Neural Inf. Process. Syst. 37, 38799–38819 (2025).
- Jiao, R., Huang, W., Liu, Y., Zhao, D. & Liu, Y. Space group constrained crystal generation. Preprint at https://arxiv.org/abs/2402.03992 (2024). · arXiv
- Ruple, L., Torresi, L., Schopmans, H. & Friederich, P. Symmetry-aware Bayesian flow networks for crystal generation. Preprint at https://arxiv.org/abs/2502.03146 (2025). · arXiv
- Kelvinius, F. E. et al. WyckoffDiff—a generative diffusion model for crystal symmetry. Preprint at https://arxiv.org/abs/2502.06485 (2025). · arXiv
- Levy, D. et al. SymmCD: symmetry-preserving crystal generation with diffusion models. Preprint at https://arxiv.org/abs/2502.03638 (2025). · arXiv
- Lipman, Y., Chen, R. T., Ben-Hamu, H., Nickel, M. & Le, M. Flow matching for generative modeling. Preprint at https://arxiv.org/abs/2210.02747 (2022). · arXiv
- Luo, X. et al. CrystalFlow: a flow-based generative model for crystalline materials. Preprint at https://arxiv.org/abs/2412.11693 (2024). · arXiv
- Vaswani, A. Attention is all you need. Adv. Neural Inf. Process. Syst. 30, 6000–6010 (2017).
- Radford, A. et al. Language models are unsupervised multitask learners. OpenAI Blog 1, 9 (2019).
- Achiam, J. et al. GPT-4 technical report. Preprint at https://arxiv.org/abs/2303.08774 (2023). · arXiv
- Gruver, N. et al. Fine-tuned language models generate stable inorganic materials as text. Preprint at https://arxiv.org/abs/2402.04379 (2024). · arXiv
- Jia, S., Zhang, C. & Fung, V. LLMatDesign: autonomous materials discovery with large language models. Preprint at https://arxiv.org/abs/2406.13163 (2024). · arXiv
- Ansari, M., Watchorn, J., Brown, C. E. & Brown, J. S. dZiner: rational inverse design of materials with AI agents. Preprint at https://arxiv.org/abs/2410.03963 (2024). · arXiv
- Chen, Z. et al. ScienceAgentBench: toward rigorous assessment of language agents for data-driven scientific discovery. Preprint at https://arxiv.org/abs/2410.05080 (2024). · arXiv
- Jaech, A. et al. OpenAI o1 system card. Preprint at https://arxiv.org/abs/2412.16720 (2024). · arXiv
- Inizan, T. J. et al. System of agentic AI for the discovery of metal–organic frameworks. Preprint at https://arxiv.org/abs/2504.14110 (2025). · arXiv
- Ortega-Ochoa, R., Aspuru-Guzik, A., Vegge, T. & Buonassisi, T. A tomographic interpretation of structure–property relations for materials discovery. Preprint at https://arxiv.org/abs/2501.18163 (2025). · arXiv
- Antoniuk, E. R. et al. Active learning enables extrapolation in molecular generative models. Preprint at https://arxiv.org/abs/2501.02059 (2025). · arXiv
- Riebesell, J. et al. Matbench discovery—a framework to evaluate machine learning crystal stability predictions. Preprint at https://arxiv.org/abs/2308.14920 (2023). · arXiv
Cited by
Peivaste, I., Belouettar, S., Mercuri, F., Fantuzzi, N., Dehghani, H., Izadi, R., Ibrahim, H., Lengiewicz, J., Belouettar-Mathis, M., Bendine, K., et al. (2025) · Composite Structures · doi:10.1016/j.compstruct.2025.119419 · Cited by 54
AI‐Driven Defect Engineering for Advanced Thermoelectric Materials
Fu, C., Cheng, M., Hung, N. T., Rha, E., Chen, Z., Okabe, R., Carrizales, D. C., Mandal, M., Cheng, Y. Q., & Li, M. (2025) · Advanced Materials · doi:10.1002/adma.202505642 · Cited by 47
Self-citationLin, W., Jin, L., Jiang, Z., Cai, M., Lai, Z., Yurchenko, D., Yan, B., Zhou, S., Novoselov, K. S., Liao, W., et al. (2026) · Advanced Functional Materials · doi:10.1002/adfm.75070 · Cited by 45
Machine learning driven design of polymer dielectrics for high temperature capacitive energy storage
Liu, D., Cheng, R., Li, Q., Wang, F., Li, Y., Yang, H., Shen, M., Zhou, Q., He, J., Li, Q., et al. (2026) · Energy & Environmental Science · doi:10.1039/d6ee01953f · Cited by 24
Chinnalagu, D. K., Sathya, A. T., Sahoo, S., & Chakraborty, B. (2026) · Journal of Materials Chemistry A · doi:10.1039/d6ta00255b · Cited by 21
Metadata
- Authors
- Cheng, M., Fu, C.-L., Okabe, R., Chotrattanapituk, A., Boonkird, A., Hung, N. T., & Li, M.
- Affiliations
- Massachusetts Institute of Technology (US)
- Frontier Research Institute for Interdisciplinary Sciences, Tohoku University (JP)
- Corresponding author
- Cheng, Mouyang
- Li, Mingda
- ORCID
- Venue
- Nature Materials
- Published
- 2026-01-02
- Volume
- 25
- Issue
- 2
- Pages
- 174-190
- ISSN
- 1476-1122 (ISSN-L)
- 1476-4660
- OpenAlex
- W7118011688
- Cited by
- 109
- References
- 96
- Citations by year
- 2026: 108
- FWCI
- 35.48
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- www.springernature.com/gp/researchers/text-and-data-mining (tdm)
- www.springernature.com/gp/researchers/text-and-data-mining (vor)
- Locations
- Nature Materials (publishedVersion)
- PubMed (publishedVersion)
- Funders
- U.S. Department of Energy (DE-SC0021940)
- National Science Foundation (ITE-2345084)
- Topics
- Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
- Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
- Physical Sciences › Materials Science › General Materials Science › Material Selection and Properties
- Keywords
- materials design, materials discovery, inverse design, deep generative models, reinforcement learning, evolutionary algorithms, machine learning, high-throughput screening, functional materials
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.