DPA4 is a new SE(3)-equivariant interatomic potential with EMFA SO(2) convolution that sets new accuracy-cost records on Matbench Discovery and SPICE benchmarks using fewer parameters than prior models.
Zhang, Phase Diagram of a Deep Potential Water Model, Phys
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
physics.chem-ph 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Prior revPBE-D3 DFT simulations of liquid water showed >20% variation in diffusion and 10% in density; new ML-potential-enabled benchmarks resolve these to consistent values across six codes, attributing prior errors to basis-set and sampling issues.
citing papers explorer
-
DPA4: Pushing the Accuracy-Cost Frontier of Interatomic Potentials with EMFA SO(2) Convolution
DPA4 is a new SE(3)-equivariant interatomic potential with EMFA SO(2) convolution that sets new accuracy-cost records on Matbench Discovery and SPICE benchmarks using fewer parameters than prior models.
-
How reproducible are first-principles simulations of liquid water?
Prior revPBE-D3 DFT simulations of liquid water showed >20% variation in diffusion and 10% in density; new ML-potential-enabled benchmarks resolve these to consistent values across six codes, attributing prior errors to basis-set and sampling issues.