density functional theory
PulseAugur coverage of density functional theory — every cluster mentioning density functional theory across labs, papers, and developer communities, ranked by signal.
8 day(s) with sentiment data
-
New framework improves MLIPs for electrostatic interactions
Researchers have developed a new framework for machine learning interatomic potentials (MLIPs) that better accounts for electrostatic effects. This framework views existing models as coarse-grained approximations of den…
-
New ontology standardizes machine learning interatomic potentials
Researchers have developed a new ontology, the MLIPs ontology, to standardize the description of machine learning interatomic potentials (MLIPs). This OWL 2 DL ontology aims to address the scattered metadata issue in th…
-
New AI framework accelerates catalyst design for targeted properties
Researchers have developed a new framework called Catalyst Diffusion Transformer (CatDiT) for designing heterogeneous catalysts. This model can generate novel and valid catalyst structures, including intermetallic alloy…
-
Flash EQ-Linear accelerates equivariant neural networks with Fourier transforms
Researchers have developed Flash EQ-Linear, a novel algorithm designed to significantly accelerate equivariant linear layers in neural networks. By leveraging the Fourier convolution theorem and the conjugate symmetry o…
-
New AI system deciphers organic structures from spectroscopic data
Researchers have developed a new hypothesis-refinement paradigm to determine organic molecular structures from spectroscopic data. This approach integrates spectral evidence with large-scale molecular priors, addressing…
-
OLEDLM: New Language Model for OLED Molecular Design
Researchers have developed OLEDLM, a novel language model specifically designed for the design of organic light-emitting diode (OLED) molecules. This model utilizes a LLaMA-style transformer architecture as a foundation…
-
Graph Neural Networks accelerate catalyst design for graphene quantum dots
Researchers have developed a novel framework utilizing graph neural networks (GNNs) to significantly accelerate the exploration of transition metal adsorption on graphene quantum dots (GQDs). This GNN-based model, named…
-
Machine learning poised to revolutionize quantum chemistry research
A new position paper argues that machine learning is the most promising direction for advancing quantum chemistry. The paper posits that traditional methods like density functional theory and wavefunction methods are re…
-
DeepCormack algorithms accelerate material Fermi surface studies
Researchers have developed DeepCormack, a novel set of data-driven algorithms designed to improve the reconstruction of 3D two-photon momentum density (TPMD) for material Fermi surface studies. This method integrates de…
-
New LLPR method boosts MLFF accuracy with fewer labels · 2 sources tracked
Researchers have developed a new active learning workflow called Last-layer-projection regression (LLPR) to improve the efficiency of training and fine-tuning machine-learning force fields (MLFFs). LLPR acts as a cost-e…
-
New ML potential TWIN models biomolecular systems with ab initio accuracy
Researchers have developed the Transferable Water Implicit Network (TWIN), a new machine learning potential for modeling biomolecular systems in aqueous environments. Unlike previous models that relied on empirical forc…
-
TWIN machine learning model accelerates biomolecular simulations with ab initio accuracy
Researchers have developed the Transferable Water Implicit Network (TWIN), a new machine learning interatomic potential (MLP) that significantly speeds up atomistic modeling for biomolecular systems. Unlike previous imp…
-
EquiFiLM enhances AI force fields for electronic state changes
Researchers have developed EquiFiLM, a novel extension for foundation machine learning force fields (MLFFs) that enables them to handle externally induced changes to electronic states. This method uses a lightweight, pe…
-
New framework speeds up neural network training for enzyme catalysis
Researchers have developed Enerzyme, a new framework designed to make training neural network potentials (NNPs) more efficient for studying enzyme catalysis. This framework addresses the computational demands of quantum…
-
New research explores weight-space geometry of AI reasoning distillation methods
A new research paper analyzes the geometric properties of weight updates across various offline reinforcement learning methods used for distilling reasoning capabilities into smaller AI models. The study trained six dif…
-
AdsMind system uses AI agents to accelerate catalyst discovery
Researchers have developed AdsMind, a novel multi-agent system designed to accelerate the discovery of adsorption configurations on heterogeneous catalyst surfaces. This closed-loop framework integrates machine learning…
-
New method extracts electrostatics from AI potentials
Researchers have developed a method called Latent Ewald Summation (LES) to extract electrostatic properties from foundation machine learning interatomic potentials (MLIPs). This technique allows for the creation of more…
-
New framework enhances reliability in AI-driven materials discovery
Researchers have developed InvDesMobility, a novel framework designed to enhance the reliability and auditability of closed-loop materials discovery. This system integrates automated density functional theory (DFT) calc…
-
New neural operator accelerates density functional theory calculations
Researchers have developed HamEvo, a novel neural operator designed to accelerate density functional theory (DFT) calculations by predicting Kohn-Sham Hamiltonians. This method achieves significant error reductions of 3…
-
New Benchmark Reveals AI Models Struggle with Crystal Stability
Researchers have introduced PhononBench, a new benchmark designed to evaluate the dynamical stability of AI-generated crystalline materials. This benchmark utilizes the MatterSim interatomic potential for efficient phon…