CrystalFlow
PulseAugur coverage of CrystalFlow — every cluster mentioning CrystalFlow across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
TCSP3.0 advances crystal structure prediction with improved accuracy and speed
A new method called TCSP3.0 has been developed for predicting crystal structures, significantly improving upon its predecessor, TCSP2.0. This enhanced approach addresses issues such as incorrect compositions and corrupt…
-
New generative model uFlowCSP drastically speeds up crystal structure prediction
Researchers have developed uFlowCSP, a novel generative model for crystal structure prediction that significantly accelerates the inference process. Unlike previous models that require thousands of sequential evaluation…
-
TFMat framework uses text-guided flow matching for efficient crystal structure generation
Researchers have developed TFMat, a novel framework that uses text-guided flow matching to generate crystal structures more efficiently. This method allows for the incorporation of structured materials language, such as…
-
AI uses language to guide crystal and molecular structure generation
Two new research papers introduce novel methods for structure generation using language-informed flow matching. The first, TFMat, uses structured text to guide crystal structure generation, improving match rates and ali…
-
New AI workflow speeds up materials discovery with surrogate-guided generation
Researchers have developed a new workflow for materials design that uses a Gaussian process surrogate to efficiently guide generative models. This approach significantly reduces the need for costly property evaluations …