2. High-resolution Image Reconstruction on Supercomputers
Chen Peng +4 more
openalex +2 more sources
TopoMAS: Large Language Model Driven Topological Materials Multi‐Agent System
TopoMAS is an interactive multi‐agent framework that revolutionizes topological materials discovery through human–AI collaborative intelligence. The system integrates natural language processing, knowledge retrieval from literature and databases, crystal structure generation, and automated first‐principles calculations within a unified workflow.
Baohua Zhang +5 more
wiley +1 more source
An Energy-Efficient Scheduling Algorithm for Shared Facility Supercomputer Centers [PDF]
E. A. Kiselev +2 more
openalex +1 more source
Molecular dynamics analysis of novel statin analogs shows that binding induces superior stabilization of HMG‐CoA reductase. As shown by the solvent‐accessible surface area profile, ligand‐induced rigidity offers a new, effective strategy for drug design. Cardiovascular diseases remain a leading cause of global mortality.
Yoshua B. Mtulo +4 more
wiley +1 more source
Development of seismic tomography software for hybrid supercomputers
A. А. Nikitin +2 more
openalex +1 more source
Characteristics of diabatically influenced cyclones with high wind damage potential in Europe
Diabatic processes contribute, on average, 26% to the intensification of European winter storms, with diabatically driven cyclones exhibiting steeper deepening rates, stronger wind gusts, and increased precipitation. These storms are linked to enhanced warm conveyor belt (WCB) activity and develop in a warmer environment with elevated lower ...
Svenja Christ +2 more
wiley +1 more source
Carbon Speciation and Solubility in Silicate Melts
This book is Open Access. A digital copy can be downloaded for free from Wiley Online Library.
Explores the behavior of carbon in minerals, melts, and fluids under extreme conditions
Carbon trapped in diamonds and carbonate-bearing rocks in subduction zones are examples of the continuing exchange of substantial carbon ...
Natalia Solomatova +2 more
wiley +1 more source
Polar‐low track prediction using machine‐learning methods
Machine‐learning models are developed to produce reliable and efficient forecasts of polar‐low (PL) trajectories 12 hours ahead. A temporal model (RLSTM) benefiting from the rolling‐forecast strategy, improves overall prediction accuracy and is suitable for quick experimentation, while a spatiotemporal model (PL‐UNet), incorporating both historical and
Ziying Yang +4 more
wiley +1 more source

