Hugging Face Trending Papers
Risk-Controlled DFT Selection for Machine-Learned Interatomic Potentials
Read the original on Hugging Face Trending Papers →The Flow has not summarised this story yet — read it at Hugging Face Trending Papers.
Related stories
arXiv Machine Learning
5d ago
Risk-Controlled DFT Selection for Machine-Learned Interatomic Potentials
arXiv:2610.19630v1 Announce Type: cross Abstract: Automated materials discovery requires evaluating large pools of candidate configurations, with density functional theory (DFT) as a major throughput...
arXiv Machine Learning
Jul 17
Full-data accuracy with fewer labels for training and fine-tuning machine-learning force fields
arXiv:2607. 14486v1 Announce Type: cross Abstract: Machine-learning force fields (MLFFs) are reliable only near their training distribution, making efficient construction of diverse training sets a major bottleneck for both train-from-scratch and foundation fine-tuning workflows.
Hugging Face Trending Papers
Oct 7
Physics-Aligned Electronic Ground-State Learning Improves Generalization
Machine-learned interatomic potentials (MLIPs) excel at in-distribution tasks, accelerating drug and material development, yet they struggle to generalize out-of-distribution. We propose to push the c...
arXiv AI
Oct 5
Equivariant Flow Matching for Electron Density Prediction
The paper introduces OrbFlow, an SE(3)-equivariant generative model that predicts Gaussian-type orbital coefficients for electron density using flow matching. By training with a two-phase trajectory curriculum, OrbFlow improves accuracy on QM9 and MD benchmarks, reducing density errors significantly compared to previous models. The predicted densities also accelerate self-consistent field procedures, cutting iterations by up to 68% and accurately reproducing dipole and quadrupole moments without additional SCF calculations.
arXiv Machine Learning
Oct 8
Physics-Aligned Electronic Ground-State Learning Improves Generalization
The paper introduces Physics‑Aligned Electronic Ground‑State Models (GSMs) that sit between machine‑learned interatomic potentials and Kohn‑Sham density functional theory in terms of computational cost and accuracy. By aligning GSM learning objectives with KS‑DFT equations and enforcing physical constraints, the authors develop new training techniques—OrthoNormal‑Loss (ON‑Loss), Grassmann Restricted Occupied‑Orbital Training (GROOT), and Residual Optimal‑gauge Conditioning‑aware KS‑Eq. Training (ROCKET)—which dramatically reduce energy and force errors in size‑extrapolation tests from QM9 to QM40. The models also achieve near‑chemical‑accuracy predictions on reactive chemistry benchmarks and demonstrate efficient label‑free self‑consistency fine‑tuning. whyItMatters":"The work advances the cost‑accuracy frontier for electronic structure predictions, enabling more reliable out‑of‑distribution generalization in drug and material discovery."
arXiv Statistics ML
Sep 28
Multivariate conformal uncertainty propagation in multitask atomistic simulation: Successes and pitfalls
The paper explores multivariate conformal uncertainty propagation for multitask atomistic simulations, introducing methods such as Bonferroni‑corrected hyperrectangles, hyperellipsoidal sets based on Mahalanobis distance, and custom loss functions within conformal risk control. It applies these techniques to calibrate predictions of energies, forces, and stresses, then propagates the resulting uncertainty sets to downstream quantities like elastic constants and vacancy formation energies. The study emphasizes how incorporating correlation predictions can capture symmetries and error cancellation, and discusses the interaction between computational protocols and conformal guarantees.