XRDiff: Crystal Structure Prediction from Powder X-Ray Diffraction Data Using Diffusion Models (opens in new tab)
Determining the crystal structure of a material from its powder X-ray diffraction (PXRD) pattern is a central challenge in materials science. PXRD is an accessible and widely used characterization technique, yet recovering the atomic structure from diffraction data requires solving an underdetermined inverse problem due to the loss of phase information. Generative modeling can provide a prior over atomic structure and learn the mapping from PX...
Read the original article