Continuous discovery of novel 2D materials via dual active learning-driven generative models
Open Access
- 12 February 2026
- journal article
- research article
- Published by Oxford University Press (OUP) in National Science Review
- Vol. 13 (7) , nwag101
Abstract
Generative artificial intelligence is transforming materials discovery by creating unexplored candidates. However, such models are trapped in historical data bias, particularly for data-scarce systems like two-dimensional (2D) materials, leading to repetitive outputs rather than genuine discoveries. Here, we introduce DuALGen, a dual active learning framework that mitigates these limitations. DuALGen couples two complementary loops to enrich data diversity and correct data bias: a generative loop that uses dynamic, multi-criteria sampling to drive exploration of the design space, and a predictive loop that samples outliers to counter distribution shift, enabling reliable evaluation of novel, previously unknown candidates. Applied to 2D materials, DuALGen uncovers >10 000 stable, distinct compounds, including thousands of high-performance candidates for electronic applications. This self-updating workflow connects generative models to uncharted chemical spaces, and offers a practical route to continuous discovery of new materials.Keywords
Funding Information
- National Key Research and Development Program of China (2022YFA1503103)
- National Natural Science Foundation of China (T2321002)
- National Natural Science Foundation of China (22373013)
- National Natural Science Foundation of China (224B2303)
- Natural Science Foundation of Jiangsu Province (BK20222007)
- Natural Science Foundation of Jiangsu Province (BK20232012)
- Fundamental Research Funds for the Central Universities (BK20233002)
This publication has 46 references indexed in Scilit:
- AI is dreaming up millions of new materials. Are they any good?Nature, 2025
- Metal-lattice-heredity synthesis of single-crystalline 2D transition metal oxidesMatter, 2024
- Universal materials model of deep-learning density functional theory HamiltonianScience Bulletin, 2024
- In Pursuit of the Exceptional: Research Directions for Machine Learning in Chemical and Materials ScienceJournal of the American Chemical Society, 2023
- On-the-fly interpretable machine learning for rapid discovery of two-dimensional ferromagnets with high Curie temperatureChem, 2022
- VASPKIT: A user-friendly interface facilitating high-throughput computing and analysis using VASP codeComputer Physics Communications, 2021
- Planar Hypercoordinate Motifs in Two-Dimensional MaterialsAccounts of Chemical Research, 2020
- Matminer: An open source toolkit for materials data miningComputational Materials Science, 2018
- Computational Screening of All Stoichiometric Inorganic MaterialsChem, 2016
- Python Materials Genomics (pymatgen): A robust, open-source python library for materials analysisComputational Materials Science, 2013