新的MLIP方法提高了准确性并实现了研究自动化

报道来源 [7]

1. arXiv cs.LG TIER_1 English(EN) · G. Laskaris, D. Morozov, D. Tarpanov, A. Seth, J. Procelewska, G. Sai Gautam, A. Sagingalieva, R. Brasher, A. Melnikov · 2026-06-08 04:00

   多目标优化与量子混合的等变深度学习原子间势

   arXiv:2602.16908v2 Announce Type: replace-cross Abstract: Allegro is a machine learning interatomic potential model designed to predict atomic properties in molecules using E(3) equivariant neural networks. When training this model, there tends to be a trade-off between accuracy …

2. arXiv cs.LG TIER_1 English(EN) · Joanna Zou, Fraser Birks, Dallas Foster, Youssef Marzouk · 2026-06-04 04:00

   Stein Kernelized Molecular Dynamics for Active Learning of Interatomic Potentials

   arXiv:2606.04100v1 Announce Type: new Abstract: Machine learning interatomic potentials (MLIPs) enable efficient and accurate atomistic simulations but depend critically on the quality and diversity of the training data. We introduce Stein kernelized molecular dynamics (SKMD), an…

3. arXiv cs.LG TIER_1 English(EN) · Zemin Xu, Chenyu Wu, Wenbo Xie, P. Hu · 2026-06-03 04:00

   用于机器学习原子间势的Cartesian-3j框架

   arXiv:2512.16882v2 Announce Type: replace-cross Abstract: Machine learning interatomic potentials (MLIPs) have brought substantial gains in the extrapolation capability in computational chemistry. However, most equivariant models are typically built with spherical tensors (STs), …

4. arXiv cs.LG TIER_1 English(EN) · Teddy Koker, Abhijeet Gangan, Mit Kotak, Jaime Marian, Tess Smidt · 2026-06-02 04:00

   PFT：用于机器学习原子间势的声子微调

   arXiv:2601.07742v4 Announce Type: replace-cross Abstract: Many materials properties depend on higher-order derivatives of the potential energy surface, yet machine learned interatomic potentials (MLIPs) trained with a standard loss on energy, force, and stress errors can exhibit …

5. arXiv cs.AI TIER_1 English(EN) · Ryan Liu, Eric Qu, Tobias Kreiman, Samuel M. Blau, Aditi S. Krishnapriyan · 2026-06-02 04:00

   从评估到设计：利用势能面光滑度指标指导机器学习原子间势能模型架构

   arXiv:2602.04861v2 Announce Type: replace-cross Abstract: Machine Learning Interatomic Potentials (MLIPs) sometimes fail to reproduce the physical smoothness of the quantum potential energy surface (PES), leading to erroneous behavior in downstream simulations that standard energ…

6. arXiv cs.LG TIER_1 English(EN) · Etinosa Osaro, Santosh Adhikari, Stamatia Zavitsanou, Kelsey Parker, Dario Rocca · 2026-06-01 04:00

   MLIPilot：LLM驱动的机器学习原子间势能自动研究

   arXiv:2605.30889v1 Announce Type: cross Abstract: Constructing production-quality machine-learned interatomic potentials (MLIPs) requires balancing accuracy, dynamical stability, and computational throughput under constraints that are not captured by a single training loss. We in…

7. arXiv stat.ML TIER_1 English(EN) · Filippo Bigi, Paolo Pegolo, Arslan Mazitov, Jonathan Schmidt, Michele Ceriotti · 2026-06-05 04:00

   突破无约束机器学习原子间势能的极限

   arXiv:2601.16195v3 Announce Type: replace-cross Abstract: Machine-learned interatomic potentials (MLIPs) are increasingly used to replace computationally demanding electronic-structure calculations to model matter at the atomic scale. The most commonly used model architectures ar…