Pith. sign in

REVIEW 17 cited by

Orb: A Fast, Scalable Neural Network Potential

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2410.22570 v1 pith:XPIHD4XT submitted 2024-10-29 cond-mat.mtrl-sci cs.LG

Mark Neumann , James Gin , Benjamin Rhodes , Steven Bennett , Zhiyi Li , Hitarth Choubisa , Arthur Hussey , Jonathan Godwin This is my paper · ORCID
classification cond-mat.mtrl-scics.LG
keywords materialsmodelpotentialsuniversalaspectsatomisticbenchmarkcarlo
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We introduce Orb, a family of universal interatomic potentials for atomistic modelling of materials. Orb models are 3-6 times faster than existing universal potentials, stable under simulation for a range of out of distribution materials and, upon release, represented a 31% reduction in error over other methods on the Matbench Discovery benchmark. We explore several aspects of foundation model development for materials, with a focus on diffusion pretraining. We evaluate Orb as a model for geometry optimization, Monte Carlo and molecular dynamics simulations.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 43 citations worldwide. Full citation record

  1. Pushing the limits of unconstrained machine-learned interatomic potentials

    physics.chem-ph 2026-01 conditional novelty 7.0 of 10

    Unconstrained non-equivariant and direct-force neural interatomic potentials scale to 730M parameters and match or beat equivariant state-of-the-art models on several atomistic benchmarks.

  2. Dyna-Mat: End-to-end benchmarking of foundation machine learning interatomic potentials in finite-temperature ensembles

    cond-mat.mtrl-sci 2026-07 conditional novelty 6.5 of 10

    Dyna-Mat-v1.0 benchmarks 15 foundation MLIPs on finite-T MD observables, finding average force-error correlation with RDF/VDOS but systematic pressure failures and near-Pareto optimality of latest cross-trained models.

  3. Implicit Machine Learning Force Fields Accelerate Molecular Dynamics Simulations

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Replacing explicit neural network stacks with self-consistent fixed-point iterations, and warm-starting the solver across timesteps, gives 2-5x cheaper molecular dynamics force evaluation at matched accuracy.

  4. From MLIPs to Microstructure: A High-Throughput Computational Framework to Design Spinodal Alloys in High-Dimensional Composition Spaces via Analytic Derivatives of CALPHAD Model Predictions

    cond-mat.mtrl-sci 2026-07 conditional novelty 6.0 of 10

    An MLIP-trained CALPHAD workflow with analytic Hessians predicts spinodal and microstructure maps in Hf-Nb-Ti-V, but quantitative mismatches remain in partitioning, misfit, and modulation direction.

  5. Rem3Di: Learning smooth, chiral 3D molecular descriptors from atomistic foundation models

    physics.chem-ph 2026-07 conditional novelty 6.0 of 10

    Rem3Di builds fixed-length, chirality-aware molecular descriptors by aggregating frozen atomistic foundation-model features with attention and a self-supervised denoising pretraining objective, matching or beating gra...

  6. Efficient Large-Scale STEM-EELS Simulations With Torched-TACAW

    cond-mat.mtrl-sci 2026-07 unverdicted novelty 6.0 of 10

    Torched-TACAW plus ORB MD and z-partitioned supercells enables near-ab-initio-quality atomic-resolution vibrational STEM-EELS for thick TiO2 models with tractable memory and data flow.

  7. Performance of universal machine learning potentials in global optimization of inorganic crystal structures

    cond-mat.mtrl-sci 2026-02 conditional novelty 6.0 of 10

    Nine universal machine-learning potentials were run in unconstrained evolutionary searches for the ground states of twelve inorganic compounds; performance ranges from near-DFT accuracy (eSEN) to essentially non-predi...

  8. Faster Molecular Dynamics with Neural Network Potentials via Distilled Multiple Time-Stepping and Non-Conservative Forces

    physics.chem-ph 2026-02 conditional novelty 6.0 of 10

    Distilled non-conservative force models in a multi-time-step integrator accelerate neural-network-potential molecular dynamics by up to 5.6x on tested systems without degrading sampling accuracy.

  9. Thermodynamic assessment of machine learning models for solid-state synthesis prediction

    cond-mat.mtrl-sci 2026-02 accept novelty 6.0 of 10

    Most machine-learning synthesizability models overpredict the likelihood of synthesizing hypothetical ternary oxides, relative to thermodynamic stability and reaction-selectivity bounds; only SynthNN tracks the thermo...

  10. Benchmarking Universal Interatomic Potentials on Zeolite Structures

    cond-mat.mtrl-sci 2025-09 accept novelty 6.0 of 10

    Universal machine-learned interatomic potentials, especially eSEN-30M-OAM, accurately reproduce DFT-level geometries and energies for zeolites, while classical universal force fields largely fail.

  11. Universal Machine Learning Potentials under Pressure

    cond-mat.mtrl-sci 2025-08 conditional novelty 6.0 of 10

    Universal machine learning interatomic potentials systematically lose accuracy under pressure up to 150 GPa, and fine-tuning on high-pressure DFT data recovers most of the lost performance.

  12. Universal Machine Learning Potential for Systems with Reduced Dimensionality

    cond-mat.mtrl-sci 2025-08 conditional novelty 6.0 of 10

    Benchmarking 11 universal machine learning interatomic potentials on a new 40,000-structure, 0D-3D dataset shows energy and geometry errors grow as dimensionality falls, with eSEN the most transferable.

  13. Distillation of atomistic foundation models across architectures and chemical domains

    physics.comp-ph 2025-06 conditional novelty 6.0 of 10

    Distillation of atomistic foundation models via synthetic data yields 10x-100x faster student potentials with near-teacher accuracy.

  14. Benchmarking Universal Machine Learning Interatomic Potentials for Real-Time Analysis of Inelastic Neutron Scattering Data

    physics.comp-ph 2025-06 conditional novelty 6.0 of 10

    Several universal machine learning interatomic potentials, especially ORB v3, MatterSim, and MACE-OFF, reach near-DFT phonon accuracy and match many experimental neutron spectra, with important caveats about test-set ...

  15. Quantum machine learning interatomic potential: Application of variational quantum algorithm

    quant-ph 2026-07 conditional novelty 5.0 of 10

    Replacing the final layer of a pretrained ANI interatomic potential with a small quantum circuit training only the circuit parameters gives slightly lower energy RMSE than a classical layer, only where the pretrained ...

  16. Leveraging neural network interatomic potentials for a foundation model of chemistry

    cond-mat.mtrl-sci 2025-06 conditional novelty 5.0 of 10

    Using embeddings from a pretrained neural network interatomic potential as features for small machine learning models gives competitive or better property predictions than end-to-end deep networks, especially with lim...

  17. Universal Machine Learning Interatomic Potentials are Ready for Solid Ion Conductors

    cond-mat.mtrl-sci 2025-02 conditional novelty 5.0 of 10

    MatterSim is the most accurate universal machine-learning potential for solid-state electrolytes across energy, force, mechanical, and lithium-diffusion benchmarks.

Pith tools