MXtalGFlow combines a canonical crystal parameterization with energy-based GFlowNet training to sample thermodynamic distributions of molecular crystals, recovering known polymorphs and predicting new competitive packing modes.
hub Mixed citations
Uma: A family of universal models for atoms
Mixed citation behavior. Most common role is background (60%).
37
Pith papers citing it
38
external citations · external index
Background
60% of classified citations
hub tools
citation-role summary
background 3
baseline 1
method 1
citation-polarity summary
representative citing papers
A self-evolving multi-agent catalysis digital twin (CatDT) matches seven gas–solid kinetic benchmarks within ~0.5–2× experiment and computationally proposes non-precious PDH candidates including Ni@ZrO2.
An E(3)-equivariant deep RL framework lets an O2 agent discover kinetically plausible diffusion and dissociation pathways in disordered Si/a-SiO2 without hand-crafted reaction coordinates or collective variables.
Kernels from pretrained MLIP latent spaces outperform standard acquisition methods in active learning for reactive chemistry, reducing required labels by 38% for energy error and 28% for force error.
PROBE recasts MLIP uncertainty quantification as selective classification by training a compact discriminative classifier on frozen per-atom backbone embeddings, yielding a reliability probability that tracks actual error better than ensemble disagreement.
A closed-loop framework jointly optimizes molecular composition and geometry in multi-component systems, demonstrated by a 30% reduction in activation barrier for a Claisen rearrangement via post-hoc validation.
SOAP and SOAP-Muon optimizers deliver faster convergence and higher final accuracy than Adam for NequIP and Allegro MLIPs, with the largest gains under partial force supervision.
CoTAR reconstructs molecular topology from atomic positions in non-reactive condensed-phase systems using a GNN-HMM hybrid, reporting 0.906 F1 on classical MD data and raising uMLIP valid-snapshot rate from 38.6% to 84.7% after few-shot fine-tuning.
A GPT-style model pretrained on 133M catalyst structures generates valid structures conditioned on categorical and continuous properties, achieving 98% structural validity and up to 4-fold screening efficiency gains.
SAP-derived features feed an orbital-based graph neural network that predicts Kohn-Sham Fock matrices, reproducing orbital energies, dipoles, density of states on QM9 and transfer integrals on organic dimers with 4.8 meV MAE on unseen cases.
Distilled compact MLIPs from transfer-learned teachers reproduce observables more reliably than same-size models trained directly and enable practical PIMD umbrella sampling of water dissociation at TiO2 interface with NQE effects matching NMR.
Physics-informed distillation from a universal MLIP plus limited CCSD(T) fine-tuning yields cm^{-1} accurate potentials for non-covalent interactions, with teacher choice strongly affecting accuracy on some systems.
Clari, a unit-cell flow matching model with pair-bias attention, generates organic crystal structures faster than OXtal while improving solve rates and supporting energy-based ranking without relaxation.
Multitask learning on linear-scaling GFN1-xTB orbital charges cuts energy MAE by 46% and data needs by 5x versus energy-only MLIPs while outperforming DFT atomic charge augmentation.
FMwC computes per-sample confidence scores for flow matching models via closed-form propagation of input-dependent multiplicative noise variance along the sampling ODE, supporting filtering, editing, and adaptive stepping.
A reweighting method with mean energy-gap approximation transfers PMFs between MLIPs to recover target reaction and activation free energies at low cost for a 601-atom Li+ transport system across DFT levels.
TSAgent automates transition state searches at DFT accuracy via an agentic loop, reaching 83% success on 100 OC20NEB examples and 70% on 10 held-out cases versus 73% for human experts.
Force-aware Neural Tangent Kernels combined with chunked acquisition provide scalable and distribution-robust active learning for MLIPs, outperforming baselines on OC20 and remaining competitive on other benchmarks.
A new benchmark finds that state-of-the-art ML interatomic potentials struggle with compositional generalization, producing errors an order of magnitude higher on unseen molecular combinations than on training-like cases.
Structural pruning of SO(3) equivariant atomistic models from large checkpoints yields 1.5-4x fewer parameters and 2.5-4x less pre-training compute than small models trained from scratch, while outperforming them on most Matbench Discovery metrics and downstream tasks.
VibroML automates remediation of dynamic instabilities in crystalline materials by combining MLIPs with genetic algorithms for polymorph search, finite-temperature MD validation, and compositional alloying to yield stable structures from databases like Alexandria.
PET-UAFD ensemble of ML potentials, calibrated on experimental cohesive energies and moduli, matches experimental accuracy on liquid properties and supplies uncertainty estimates via the PET-EXP protocol.
Suiren-1.0 is a family of three molecular foundation models (Base, Dimer, ConfAvg) pre-trained on 70M+ DFT samples and distilled to achieve claimed state-of-the-art performance on quantum property prediction tasks from 2D inputs.
UBio-MolFM achieves ab initio-level fidelity on large out-of-distribution biomolecular systems using a new multi-fidelity dataset, E2Former-V2 architecture, and three-stage curriculum learning.
citing papers explorer
-
A Distributional Framework for Generative Modeling of Molecular Crystals
MXtalGFlow combines a canonical crystal parameterization with energy-based GFlowNet training to sample thermodynamic distributions of molecular crystals, recovering known polymorphs and predicting new competitive packing modes.
-
Autonomous heterogeneous catalyst discovery with a self-evolving multi-agent digital twin
A self-evolving multi-agent catalysis digital twin (CatDT) matches seven gas–solid kinetic benchmarks within ~0.5–2× experiment and computationally proposes non-precious PDH candidates including Ni@ZrO2.
-
Bridging Atomistic Simulation and Experimental Processing Timescales with Goal-Directed Deep Reinforcement Learning
An E(3)-equivariant deep RL framework lets an O2 agent discover kinetically plausible diffusion and dissociation pathways in disordered Si/a-SiO2 without hand-crafted reaction coordinates or collective variables.
-
Pretrained Model Representations as Acquisition Signals for Active Learning of MLIPs
Kernels from pretrained MLIP latent spaces outperform standard acquisition methods in active learning for reactive chemistry, reducing required labels by 38% for energy error and 28% for force error.
-
Knowing when to trust machine-learned interatomic potentials
PROBE recasts MLIP uncertainty quantification as selective classification by training a compact discriminative classifier on frozen per-atom backbone embeddings, yielding a reliability probability that tracks actual error better than ensemble disagreement.
-
Hierarchical generative modeling for the design of multi-component systems
A closed-loop framework jointly optimizes molecular composition and geometry in multi-component systems, demonstrated by a 30% reduction in activation barrier for a Claisen rearrangement via post-hoc validation.
-
Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials
SOAP and SOAP-Muon optimizers deliver faster convergence and higher final accuracy than Adam for NequIP and Allegro MLIPs, with the largest gains under partial force supervision.
-
CoTAR: Topology and Atomic State Reconstruction in Condensed Phases
CoTAR reconstructs molecular topology from atomic positions in non-reactive condensed-phase systems using a GNN-HMM hybrid, reporting 0.906 F1 on classical MD data and raising uMLIP valid-snapshot rate from 38.6% to 84.7% after few-shot fine-tuning.
-
Toward Controllable Catalyst Inverse Design via Large-Scale Autoregressive Pretraining
A GPT-style model pretrained on 133M catalyst structures generates valid structures conditioned on categorical and continuous properties, achieving 98% structural validity and up to 4-fold screening efficiency gains.
-
Transferable Machine Learning of Electronic Hamiltonians with Superposition-of-Atomic-Potentials Features
SAP-derived features feed an orbital-based graph neural network that predicts Kohn-Sham Fock matrices, reproducing orbital energies, dipoles, density of states on QM9 and transfer integrals on organic dimers with 4.8 meV MAE on unseen cases.
-
Distilling first-principles accuracy into compact machine learning potentials for condensed-phase chemistry
Distilled compact MLIPs from transfer-learned teachers reproduce observables more reliably than same-size models trained directly and enable practical PIMD umbrella sampling of water dissociation at TiO2 interface with NQE effects matching NMR.
-
Non-covalent Interactions at cm$^{-1}$ Accuracy: Data Efficient Physics-Informed Distillation for Machine Learning Interatomic Potentials
Physics-informed distillation from a universal MLIP plus limited CCSD(T) fine-tuning yields cm^{-1} accurate potentials for non-covalent interactions, with teacher choice strongly affecting accuracy on some systems.
-
Fast Organic Crystal Structure Prediction with Unit Cell Flow Matching
Clari, a unit-cell flow matching model with pair-bias attention, generates organic crystal structures faster than OXtal while improving solve rates and supporting energy-based ranking without relaxation.
-
Multitask learning with semiempirical orbital charges enables sample-efficient MLIPs
Multitask learning on linear-scaling GFN1-xTB orbital charges cuts energy MAE by 46% and data needs by 5x versus energy-only MLIPs while outperforming DFT atomic charge augmentation.
-
Flowing with Confidence
FMwC computes per-sample confidence scores for flow matching models via closed-form propagation of input-dependent multiplicative noise variance along the sampling ODE, supporting filtering, editing, and adaptive stepping.
-
Reweighting free energy profiles between universal machine learning interatomic potentials for fast consensus building
A reweighting method with mean energy-gap approximation transfers PMFs between MLIPs to recover target reaction and activation free energies at low cost for a 601-atom Li+ transport system across DFT levels.
-
TSAgent: An Agentic Workflow for Autonomous Transition State Search
TSAgent automates transition state searches at DFT accuracy via an agentic loop, reaching 83% success on 100 OC20NEB examples and 70% on 10 held-out cases versus 73% for human experts.
-
Force-Aware Neural Tangent Kernels for Scalable and Robust Active Learning of MLIPs
Force-aware Neural Tangent Kernels combined with chunked acquisition provide scalable and distribution-robust active learning for MLIPs, outperforming baselines on OC20 and remaining competitive on other benchmarks.
-
Benchmarking Compositional Generalisation for Machine Learning Interatomic Potentials
A new benchmark finds that state-of-the-art ML interatomic potentials struggle with compositional generalization, producing errors an order of magnitude higher on unseen molecular combinations than on training-like cases.
-
Compact SO(3) Equivariant Atomistic Foundation Models via Structural Pruning
Structural pruning of SO(3) equivariant atomistic models from large checkpoints yields 1.5-4x fewer parameters and 2.5-4x less pre-training compute than small models trained from scratch, while outperforming them on most Matbench Discovery metrics and downstream tasks.
-
VibroML: an automated toolkit for high-throughput vibrational analysis and dynamic instability remediation of crystalline materials using machine-learned potentials
VibroML automates remediation of dynamic instabilities in crystalline materials by combining MLIPs with genetic algorithms for polymorph search, finite-temperature MD validation, and compositional alloying to yield stable structures from databases like Alexandria.
-
Errors that matter: Uncertainty-aware universal machine-learning potentials calibrated on experiments
PET-UAFD ensemble of ML potentials, calibrated on experimental cohesive energies and moduli, matches experimental accuracy on liquid properties and supplies uncertainty estimates via the PET-EXP protocol.
-
Suiren-1.0 Technical Report: A Family of Molecular Foundation Models
Suiren-1.0 is a family of three molecular foundation models (Base, Dimer, ConfAvg) pre-trained on 70M+ DFT samples and distilled to achieve claimed state-of-the-art performance on quantum property prediction tasks from 2D inputs.
-
UBio-MolFM: A Universal Molecular Foundation Model for Bio-Systems
UBio-MolFM achieves ab initio-level fidelity on large out-of-distribution biomolecular systems using a new multi-fidelity dataset, E2Former-V2 architecture, and three-stage curriculum learning.
-
Efficient Large-Scale STEM-EELS Simulations With Torched-TACAW
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.
-
Fine-Tuning a Universal Machine-Learned Interatomic Potential for Oxygen Plasma Interactions with WS$_2$
A pretrained universal MLIP without spin or Hubbard-U corrections predicts the same chemisorbed S removal and O uptake under 15 eV O+/O2+ bombardment of WS2 as after three rounds of PBE+D3+U+spin fine-tuning, which still reduces energy/force MAE to 4.5 meV/atom and 0.076 eV/A.
-
Fine-tuning MLIP foundation models: strategies for accuracy and transferability
Systematic tests show naive fine-tuning excels for single-task accuracy while multihead replay best preserves out-of-distribution robustness in MLIP adaptation.
-
Robust AI-Driven Discovery of Electronic Metal Phosphide Semiconductors
An AI-driven workflow using Wyckoff substitution, MatterGen generation, DPA3 ML potentials, and DFT identifies 3574 new stable phosphides including 196 semiconductors and screens 30 optoelectronic plus 26 thermoelectric candidates.
-
Harnessing AtomisticSkills for Agentic Atomistic Research
AtomisticSkills is a new harness framework with 100+ human-curated skills that lets general AI agents perform atomistic research tasks including simulations, screening, and analysis, shown on electrolyte design, CO2 capture, drug screening, and catalyst tasks.
-
PRISMat: Policy-Driven, Permutation-Invariant Autoregressive Material Generation
PRISMat generates crystal slabs with mean absolute errors of 0.188 eV/A² for cleavage energy and 2.79 eV for work function, reducing error by 4× versus the next best model while using less inference time.
-
Systematic Fine-Tuning of MACE Interatomic Potentials for Catalysis
Fine-tuned MACE MLIPs achieve lower mean absolute errors on catalytic reaction energies and barriers than from-scratch models, with a large fine-tuned model performing best on both metallic and oxide systems including out-of-distribution cases.
-
Benchmarking Chemically Scalable Machine-Learning Interatomic Potentials for Large-Scale Simulations of Multicomponent Alloys
GRACE-FS is generally more accurate and transferable while UNEP-v1 is much faster and supports uncertainty-aware multi-million-atom shock simulations of multicomponent alloys.
-
Accuracy and Efficiency Benchmarks of Pretrained Machine Learning Potentials for Molecular Simulations
Benchmarks of 15 MLIPs show parameter count and training set size correlate with accuracy, architecture drives speed and memory, and explicit Coulomb terms provide no benefit.
-
Comparing the latent features of universal machine-learning interatomic potentials
Different uMLIPs encode chemical space in distinct ways, with high cross-model feature reconstruction errors, and fine-tuning preserves strong pre-training bias in the latent features.
-
Machine Learning Interatomic Potentials: Advancing Open-Source Software for Efficient and Scalable Molecular Simulation
mlip v2 is a new software release that integrates API redesign, e3j backend, eSEN model, improved charge modeling, and expanded simulation capabilities to support larger-scale molecular modeling.
-
Assessing foundational atomistic models for iron alloys under Earth's core conditions
Foundational atomistic models reproduce some structural and dynamical properties of iron alloys under core conditions but none consistently match first-principles benchmarks due to missing explicit treatment of thermal electronic excitations.
-
Six Open Questions in Machine-Learned Interatomic Potential Foundation Models
This perspective article develops a definition of foundational MLIPs and poses six open questions that the authors believe will define future research in machine-learned interatomic potentials.