DPA-2: a large atomic model as a multi-task learner

The rapid advancements in artificial intelligence (AI) are catalyzing transformative changes in atomic modeling, simulation, and design. AI-driven potential energy models have demonstrated the capability to conduct large-scale, long-duration simulations with the accuracy of ab initio electronic structure methods. However, the model generation process remains a bottleneck for large-scale applications. We propose a shift towards a model-centric ecosystem, wherein a large atomic model (LAM), pre-trained across multiple disciplines, can be efficiently fine-tuned and distilled for various downstream tasks, thereby establishing a new framework for molecular modeling. In this study, we introduce the DPA-2…

• NS
  National Science Foundation

  Award: 2209718
• UD
  U.S. Department of Energy
• NN
  National Natural Science Foundation of China

  Awards: 92270001, 91961204, 22225302, 12122103, 12288101, 12122401, 12074007, 22173032, 12034009, 22222303, 12135002
• NI
  National Institutes of Health

  Award: GM107485
• NS
  National Science Fund for Distinguished Young Scholars

  Award: 22225302
• NS
  Natural Science Foundation of Zhejiang Province

  Award: 2022XHSJJ006