arXiv:2606. 08148v1 Announce Type: cross Abstract: Interatomic potentials (IPs) enable large-scale atomistic simulations beyond the reach of first-principles methods, but their predictive reliability depends critically on the selection of training data, quantified uncertainty, and model expressiveness.
By Yonatan Kurniawan (Department of Physics and Astronomy, Brigham Young University, Provo, UT, USA), Logan D. Williams (Lawrence Livermore National Laboratory, Livermore, CA, USA), Amit Samanta (Lawrence Livermore National Laboratory, Livermore, CA, USA), Ilia Nikiforov (Department of Aerospace Engineering and Mechanics, University of Minnesota, Minneapolis, MN, USA), Daniel Schwalbe-Koda (Department of Materials Science and Engineering, University of California, Los Angeles, CA, USA), Mark K. Transtrum (Cross Stream Consulting, Springville, UT, USA), Ellad B. Tadmor (Department of Aerospace Engineering and Mechanics, University of Minnesota, Minneapolis, MN, USA), Vincenzo Lordi (Lawrence Livermore National Laboratory, Livermore, CA, USA), Vasily V. Bulatov (Lawrence Livermore National Laboratory, Livermore, CA, USA)
arXiv:2603. 10950v2 Announce Type: replace Abstract: Machine learning methods for identifying molecular structures from tandem mass spectra (MS/MS) have advanced rapidly, yet current approaches still exhibit significant error rates.
By Mira J\"urgens, Gaetan De Waele, Morteza Rakhshaninejad, Willem Waegeman
arXiv:2606. 30961v1 Announce Type: cross Abstract: Advances in deep learning architectures and representations have enabled ML-driven chemical property prediction, but state-of-the-art (SOTA) models have remained largely confined to independent codebases and lack support for diverse chemical species.
By Jacob W. Toney, Samir Darouich, Yiran Wang, Aaron G. Garrison, Johannes K\"astner, Heather J. Kulik
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.
By Sheng Bi, Yi-Ze Wang, Jun Cheng
The study examines how multi‑fidelity training—using abundant low‑fidelity data and scarce high‑fidelity labels—affects uncertainty decomposition in atomistic models. By training deep ensembles of mean‑variance MACE models on ANI‑1x, the authors compare high‑fidelity training from scratch with low‑fidelity pretraining followed by fine‑tuning, evaluating aleatoric, epistemic, and total uncertainty under unseen molecular systems and higher‑energy configurations. Results show that pretraining lowers prediction error and improves uncertainty‑based ranking, but does not produce well‑calibrated predictive variances, especially under energy extrapolation, and that fine‑tuning yields a more effective uncertainty‑based acquisition function.
By Matteo Tolloso, Davide Bacciu, Luigi Bonati