7 papers
Hypothesis-driven construction of mesoscopic dynamics
Zhuoyuan Li, Aiqing Zhu, Qianxiao Li
Traditional scientific modeling typically begins with fixed, instance-wise effective equations and then carries out equation-specific analysis and computation, a procedure that bec…
In-context modeling as a retrain-free paradigm for foundation models in computational science
Lingfeng Li, Zhuoyuan Li, Shun Li +4
Building models that generalize across physical systems without retraining remains a central challenge in computational science. Here we introduce In-Context Modeling (ICM), a retr…
ABACUS: An Electronic Structure Analysis Package for the AI Era
Weiqing Zhou, Daye Zheng, Qianrui Liu +55
ABACUS (Atomic-orbital Based Ab-initio Computation at USTC) is an open-source software for first-principles electronic structure calculations and molecular dynamics simulations. It…
Active Learning for Conditional Inverse Design with Crystal Generation and Foundation Atomic Models
Zhuoyuan Li, Siyu Liu, Beilin Ye +2
Artificial intelligence (AI) is transforming materials science, enabling both theoretical advancements and accelerated materials discovery. Recent progress in crystal generation mo…
Large Language Models for Material Property Predictions: elastic constant tensor prediction and materials design
Siyu Liu, Tongqi Wen, Beilin Ye +2
Efficient and accurate prediction of material properties is critical for advancing materials design and applications. The rapid-evolution of large language models (LLMs) presents a…
Dflow, a Python framework for constructing cloud-native AI-for-Science workflows
Xinzijian Liu, Yanbo Han, Zhuoyuan Li +15
In the AI-for-science era, scientific computing scenarios such as concurrent learning and high-throughput computing demand a new generation of infrastructure that supports scalable…