Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Matter

Designing complex concentrated alloys with quantum machine learning and language modeling

Zongrui Pei, Yilun Gong, Xianglin Liu, Junqi Yin

Published Jun 21, 2024• 13 citations
Publisher
DOI
10.1016/j.matt.2024.05.035

Related papers

Quantum Computers and Hybrid Machine Learning Models for the Discovery of Lightweight Structural Alloys

S. V. MORE S. K. PANCHAL, Megh Raval, Raj Pandya +1 more

The minerals, metals & materials series•Jan 1, 2026•0 citations

Quantum machine-learning phase prediction of high-entropy alloys

Payden Brown, Houlong L. Zhuang

Materials Today•Mar 1, 2023•44 citations

UML-based Modeling of Quantum Machine Learning Systems: A Software Engineering Approach

Kartik Sharma, VidhyaLakshmi. M, Kumaran K +1 more

Quantum Machine learning (QML) is the combination of computational power of quantum computing and machine learning, which forms a highly complex system design. The hybrid nature of quantum-classical interaction as well as hardware-specific constraints in the field of quantum computing, poses unique challenges in understanding and designing quantum machine learning systems. This Study presents a structured approach in designing and modeling quantum machine learning systems using widely used software engineering practices by utilizing the Unified Modeling Language (UML) diagrams. Different UML diagrams, including use case diagram, sequence diagram, and Class diagram, are incorporated in this study to capture essential components and workflows in the quantum machine learning system and their applications. A requirement engineering lifecycle diagram is also included in this study to incorporate and understand how software engineering practices are applied to quantum computing. Through various UML diagrams, this study aims to illustrate how these models enhance modularity and development planning. To ground the modeling approach, a disease prediction system is utilized as a demonstrative example in all the modelling diagrams. This work aims to provide a foundational framework that can help and assist researchers and developers to design, communicate, and evolve quantum machine learning systems effectively and with a better and clearer vision.

International Conference on Innovative Mechanisms for Industry Applications•Sep 3, 2025•0 citations

Language Model Optimization Based on Quantum Machine Learning for Long Chinese Text Processing

Tianze Wang

Chinese language presents unique challenges in text processing due to its complex structure, including high inter-lexical dependence, flexible word order, and rich contextual expressions. These characteristics make it particularly difficult for large models to effectively handle long Chinese texts using traditional machine learning (ML) algorithms. This paper explores the application of Quantum Machine Learning (QML) as a solution to enhance the performance of large language models (LLMs) in processing long Chinese texts. By leveraging the inherent advantages of QML, such as higher processing efficiency and deeper semantic understanding, this research demonstrates through experimental comparisons that QML models significantly outperform traditional ML models in both speed and comprehension accuracy. The paper systematically introduces text processing methods using both traditional ML and QML, and presents a simplified QML model designed to enhance text processing capabilities. The findings underscore the potential of QML in optimizing the future performance of LLMs, particularly for languages with complex linguistic structures like Chinese. This research holds significant implications for advancing LLM technologies in text-heavy applications.

International Conference on Computing, Robotics and System Sciences (ICRSS)•Nov 22, 2024•0 citations

Efficient quantum algorithm for the design of complex materials: quantum circuit learning

Sota Osaki, Kazuki Hoshitani, Makoto Nakamura +2 more

Validity of the quantum circuit learning (QCL) method for the prediction of physical properties of complex materials has been explored by comparing the prediction results of Vickers hardness of the high entropy alloys with those predicted by the conventional linear and nonlinear types of machine learning methods. The linear models include linear regression (LR), ridge regression (Ridge), Bayesian ridge regression (BR), and linear support vector regression (linear_SVR) for linear models, and the nonlinear models include Gaussian process regression (GBR), support vector regression with an radial basis function kernel (rbf_SVR), random forest (RF), gradient boosting decision trees (GBDT), and multilayer perceptron neural networks (MLP). Our current examinations revealed that the QCL can efficiently predict Vickers hardness of the high entropy alloys for outside applicability domains and extrapolations even with a small number of datasets, which implies that the QCL can be used for the design of new complex materials from an early stage of development with a small dataset.

Scientific Reports•Apr 20, 2026•0 citations

Quantum Machine Learning for Materials Science

Zongrui Pei

The Materials Research Society series•Jan 1, 2026•1 citation
Quantum Machine learning (QML) is the combination of computational power of quantum computing and machine learning, which forms a highly complex system design. The hybrid nature of quantum-classical interaction as well as hardware-specific constraints in the field of quantum computing, poses unique challenges in understanding and designing quantum machine learning systems. This Study presents a structured approach in designing and modeling quantum machine learning systems using widely used software engineering practices by utilizing the Unified Modeling Language (UML) diagrams. Different UML diagrams, including use case diagram, sequence diagram, and Class diagram, are incorporated in this study to capture essential components and workflows in the quantum machine learning system and their applications. A requirement engineering lifecycle diagram is also included in this study to incorporate and understand how software engineering practices are applied to quantum computing. Through various UML diagrams, this study aims to illustrate how these models enhance modularity and development planning. To ground the modeling approach, a disease prediction system is utilized as a demonstrative example in all the modelling diagrams. This work aims to provide a foundational framework that can help and assist researchers and developers to design, communicate, and evolve quantum machine learning systems effectively and with a better and clearer vision.
Chinese language presents unique challenges in text processing due to its complex structure, including high inter-lexical dependence, flexible word order, and rich contextual expressions. These characteristics make it particularly difficult for large models to effectively handle long Chinese texts using traditional machine learning (ML) algorithms. This paper explores the application of Quantum Machine Learning (QML) as a solution to enhance the performance of large language models (LLMs) in processing long Chinese texts. By leveraging the inherent advantages of QML, such as higher processing efficiency and deeper semantic understanding, this research demonstrates through experimental comparisons that QML models significantly outperform traditional ML models in both speed and comprehension accuracy. The paper systematically introduces text processing methods using both traditional ML and QML, and presents a simplified QML model designed to enhance text processing capabilities. The findings underscore the potential of QML in optimizing the future performance of LLMs, particularly for languages with complex linguistic structures like Chinese. This research holds significant implications for advancing LLM technologies in text-heavy applications.
Validity of the quantum circuit learning (QCL) method for the prediction of physical properties of complex materials has been explored by comparing the prediction results of Vickers hardness of the high entropy alloys with those predicted by the conventional linear and nonlinear types of machine learning methods. The linear models include linear regression (LR), ridge regression (Ridge), Bayesian ridge regression (BR), and linear support vector regression (linear_SVR) for linear models, and the nonlinear models include Gaussian process regression (GBR), support vector regression with an radial basis function kernel (rbf_SVR), random forest (RF), gradient boosting decision trees (GBDT), and multilayer perceptron neural networks (MLP). Our current examinations revealed that the QCL can efficiently predict Vickers hardness of the high entropy alloys for outside applicability domains and extrapolations even with a small number of datasets, which implies that the QCL can be used for the design of new complex materials from an early stage of development with a small dataset.