pith. sign in

arxiv: 2605.15630 · v1 · pith:WAFK52M5new · submitted 2026-05-15 · ⚛️ physics.chem-ph · cond-mat.stat-mech· cs.LG

Reweighting free energy profiles between universal machine learning interatomic potentials for fast consensus building

Sauradeep Majumdar , Miguel Steiner , Johannes C. B. Dietschreit , Swagata Roy , Daniel Willimetz , Luka\v{s} Grajciar , Rafael G\'omez-Bombarelli This is my paper

Pith reviewed 2026-05-19 19:44 UTC · model grok-4.3

classification ⚛️ physics.chem-ph cond-mat.stat-mechcs.LG
keywords free energy reweightingmachine learning interatomic potentialspotential of mean forceLi ion transportnanoconfined electrolyteconsensus buildingDFT accuracyanalytical corrections
0
0 comments X p. Extension
pith:WAFK52M5 Add to your LaTeX paper What is a Pith Number? 
\usepackage{pith}
\pithnumber{WAFK52M5}

Prints a linked pith:WAFK52M5 badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

The pith

A mean energy-gap approximation reweights potential of mean force profiles from one universal MLIP to match target MLIPs even with low phase-space overlap.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a framework to reweight free energy profiles sampled with a source machine learning interatomic potential onto multiple target potentials. Traditional reweighting fails for large systems because of poor overlap between the potentials' sampled configurations. The authors show that an analytical correction based on the mean energy gap between potentials produces stable potentials of mean force that closely match what full simulations with each target potential would give. This recovers accurate reaction and activation free energies for lithium ion transport in a 601-atom nanoconfined electrolyte system. The approach works across several DFT-trained reference levels and reveals that the MLIPs form two clusters based on their training data.

Core claim

By deploying robust analytical corrections to direct exponential reweighting, the mean energy-gap approximation bypasses statistical collapse and yields a highly stable PMF that matches the target PMF, thereby recovering high-fidelity target thermodynamics at a fraction of the cost of full simulations.

What carries the argument

The mean energy-gap approximation with associated analytical corrections for reweighting PMFs between source and target MLIPs.

If this is right

  • High-fidelity target thermodynamics are recovered across multiple DFT reference levels (PBE+D3, PBE-sol, r2SCAN, r2SCAN-D4).
  • Thermodynamic analysis shows the MLIPs partition into two distinct clusters driven by training data.
  • The framework recovers reaction and activation free energies even when phase-space overlap is critically low.
  • This establishes a diagnostic protocol for affordable cross-model consensus on materials chemistry properties.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The method could extend to other large molecular systems where direct reweighting fails due to size.
  • Consensus across many MLIPs might become routine without running separate simulations for each.
  • Systematic biases might still appear in properties sensitive to rare events not captured by the mean gap.

Load-bearing premise

The mean energy-gap approximation and analytical corrections produce accurate target PMFs without systematic biases even when phase-space overlap between source and target MLIPs is critically low in a 601-atom system.

What would settle it

A full simulation using one of the target MLIPs that shows a significantly different PMF shape or free energy values compared to the reweighted result from the source MLIP.

read the original abstract

Free energy profiles serve as a fundamental bridge between microscopic atomic fluctuations and macroscopic thermodynamic observables. Estimating the free energy profile along a reaction coordinate, referred to as the potential of mean force (PMF), with density functional theory (DFT) accuracy is computationally expensive. Universal machine learning interatomic potentials (MLIPs) drastically reduce this cost, but their accuracy is strongly determined by their training data and hence can be uncertain for a given system. In this work, we present a systematic and scalable framework for reweighting PMFs, initially sampled with a single 'source' MLIP, across a representative suite of target MLIPs. Because traditional direct exponential reweighting fails for large system sizes due to low phase-space overlap between potentials, we deploy robust analytical corrections. Applying this to a complex 601-atom system of Li$^+$ transport in a nanoconfined electrolyte, we demonstrate that a mean energy-gap approximation effectively bypasses statistical collapse, producing a highly stable PMF matching the target PMF. Using this approach, we recover high-fidelity target thermodynamics across multiple DFT reference levels (PBE+D3, PBE-sol, r$^2$SCAN,r$^2$SCAN-D4) at a fraction of the computational cost of full simulations. Furthermore, thermodynamic analysis reveals that the studied MLIPs partition into two distinct clusters driven by their training data. Our reweighting framework successfully recovers target thermodynamic properties--specifically, reaction and activation free energies--even when the phase-space overlap between potentials is critically low. Ultimately, this approach establishes a vital diagnostic protocol to achieve affordable cross-model consensus on materials chemistry properties without redundant, resource-intensive simulations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript presents a reweighting framework for potential of mean force (PMF) profiles sampled with a source universal MLIP and reweighted to target MLIPs. Direct exponential reweighting is replaced by analytical corrections plus a mean energy-gap approximation to avoid statistical collapse at low phase-space overlap. The method is demonstrated on Li+ transport in a 601-atom nanoconfined electrolyte, recovering stable PMFs and thermodynamic quantities (reaction and activation free energies) that match targets across DFT levels (PBE+D3, PBE-sol, r2SCAN, r2SCAN-D4) at reduced cost. The work also reports that the MLIPs cluster into two groups according to training data.

Significance. If the central approximation is shown to be unbiased, the framework would allow rapid cross-MLIP consensus on free-energy profiles for realistic materials systems without repeating expensive sampling for each target potential. The explicit demonstration on a 601-atom system and the identification of training-data-driven clustering are concrete strengths that could inform practical MLIP usage in chemistry.

major comments (1)
  1. [Methods and results on the mean energy-gap approximation] The mean energy-gap approximation (described in the methods and results sections on reweighting) is load-bearing for the claim that target PMFs and thermodynamics are recovered without systematic bias. The manuscript provides no explicit demonstration that the neglected higher-moment contributions (variance and skewness of the energy gap) remain smaller than kT along the reaction coordinate when phase-space overlap is critically low for the 601-atom system, nor that the analytical corrections fully restore those contributions. This validation is required to support the assertion that the approximation bypasses statistical collapse reliably.
minor comments (2)
  1. [Results] Quantitative error metrics (e.g., RMSD between reweighted and target PMFs, or propagated uncertainties on free energies) are mentioned only qualitatively in the abstract and results; adding these to a table or figure would strengthen the comparison across DFT levels.
  2. [Thermodynamic analysis] The clustering of MLIPs into two groups is reported but the distance metric or linkage method used for the partition is not stated; a brief methods sentence would clarify reproducibility.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback and for recognizing the potential of our reweighting framework. We address the major comment point by point below.

read point-by-point responses

  1. Referee: The mean energy-gap approximation (described in the methods and results sections on reweighting) is load-bearing for the claim that target PMFs and thermodynamics are recovered without systematic bias. The manuscript provides no explicit demonstration that the neglected higher-moment contributions (variance and skewness of the energy gap) remain smaller than kT along the reaction coordinate when phase-space overlap is critically low for the 601-atom system, nor that the analytical corrections fully restore those contributions. This validation is required to support the assertion that the approximation bypasses statistical collapse reliably.

    Authors: We agree that an explicit validation of the mean energy-gap approximation's assumptions would strengthen the manuscript. In the original submission, the primary evidence for the approximation's validity is the close quantitative agreement between the reweighted PMFs and those obtained from independent sampling using the target MLIPs, as shown in Figure 3 and the associated thermodynamic quantities in Table 1. This agreement holds even in regimes of low phase-space overlap. However, we acknowledge the value of directly inspecting the higher moments. In the revised manuscript, we will add an analysis in the Supplementary Information that computes and plots the variance and skewness of the energy gap distribution along the reaction coordinate for the 601-atom Li+ system. We will demonstrate that these contributions are smaller than kT in the regions of interest, thereby justifying the approximation. We will also clarify the role of the analytical corrections in accounting for the mean energy shift and how they complement the approximation to reduce bias. This addition will directly address the concern about systematic bias. revision: yes

Circularity Check

0 steps flagged

Reweighting framework relies on independent mean energy-gap approximation and analytical corrections; no reduction to inputs or self-citation chain

full rationale

The paper introduces a reweighting method for PMFs between MLIPs by replacing direct exponential reweighting (which fails at low overlap in large systems) with a mean energy-gap approximation plus analytical corrections. This is presented as a methodological choice to recover target thermodynamics without redundant simulations. No quoted equations or steps show the recovered PMF or free energies reducing by construction to fitted parameters from the target itself, nor does any load-bearing premise rest solely on self-citation. The central claim remains an approximation technique whose validity is asserted via comparison to multiple DFT references, making the derivation self-contained against external benchmarks rather than tautological.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the validity of the mean energy-gap approximation for correcting low phase-space overlap and on the assumption that the chosen suite of target MLIPs is representative of DFT accuracy levels. No explicit free parameters or new entities are introduced in the abstract; the work relies on standard statistical mechanics reweighting principles.

axioms (1)
  • domain assumption Standard statistical mechanics relations for exponential reweighting of potentials of mean force remain applicable after analytical corrections.
    Invoked when stating that direct exponential reweighting fails for large systems and that corrections can restore stability.

pith-pipeline@v0.9.0 · 5876 in / 1566 out tokens · 62852 ms · 2026-05-19T19:44:52.183657+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

89 extracted references · 89 canonical work pages

  1. [1]

    Campbell, S. L. and Gear, C. W. The index of general nonlinear D A E S. Numer. M ath. 1995

    work page 1995

  2. [2]

    Slifka, M. K. and Whitton, J. L. Clinical implications of dysregulated cytokine production. J. M ol. M ed. 2000

    work page 2000

  3. [3]

    Erlebach, Andreas and. Nat. Commun. , number =

    work page

  4. [4]

    Jiancheng Yang and Qiang Zhang and Bingbing Ni and Linguo Li and Jinxian Liu and Mengdie Zhou and Qi Tian , booktitle =

    work page

  5. [5]

    Johnson, Alexander Lawrence , number =

    work page

  6. [6]

    Heard , journal =

    Chen Lei and Carlos Bornes and Oscar Bengtsson and Andreas Erlebach and Ben Slater and Lukas Grajciar and Christopher J. Heard , journal =. A machine learning approach for dynamical modelling of

    work page

  7. [7]

    Eduardo P\'. Chem. Rev. , pages =

    work page

  8. [8]

    Roy, Swagata and D. Nat. Commun. , number =

    work page

  9. [9]

    Albert P. Bart\'. Phys. Rev. B , number =

    work page

  10. [10]

    Heard and Luk\'

    Daniel Willimetz and Andreas Erlebach and Christopher J. Heard and Luk\'. Digital Discovery , volume=

    work page

  11. [11]

    Batatia, Ilyes and Benner, Philipp and Chiang, Yuan and Elena, Alin M and Kov. J. Chem. Phys. , number =

    work page

  12. [12]

    Guorui Cai and Amanda A Chen and Sharon Lin and Dong Ju Lee and Kunpeng Yu and John Holoubek and Yijie Yin and Anthony U Mu and Ying Shirley Meng and Ping Liu and others , journal =

    work page

  13. [13]

    Christophe Chipot and Andrew Pohorille , publisher =

    work page

  14. [14]

    Dietschreit, Johannes CB and Diestler, Dennis J and Hulm, Andreas and Ochsenfeld, Christian and G. J. Chem. Phys. , number =

    work page

  15. [15]

    Johannes CB Dietschreit and Dennis J Diestler and Christian Ochsenfeld , journal =

    work page

  16. [16]

    Dietschreit, Johannes CB and Diestler, Dennis J and G. J. Chem. Theory Comput. , number =

    work page

  17. [18]

    Wenxin Hu and Pengfei Li and Jia-Ning Wang and Yuanfei Xue and Yan Mo and Jun Zheng and Xiaoliang Pan and Yihan Shao and Ye Mei , journal =

    work page

  18. [19]

    Phillip S Hudson and H Lee Woodcock and Stefan Boresch , journal =

    work page

  19. [20]

    Anna S Kamenik and Stephanie M Linker and Sereina Riniker , journal =

    work page

  20. [21]

    Aaron D Kaplan and Runze Liu and Ji Qi and Tsz Wai Ko and Bowen Deng and Janosh Riebesell and Gerbrand Ceder and Kristin A Persson and Shyue Ping Ong , journal =

    work page

  21. [22]

    Gerhard K\"onig and Phillip S Hudson and Stefan Boresch and H Lee Woodcock , journal =

    work page

  22. [23]

    Shankar Kumar and John M Rosenberg and Djamal Bouzida and Robert H Swendsen and Peter A Kollman , journal =

    work page

  23. [24]

    Batatia, Ilyes and Benner, Philipp and Chiang, Yuan and Elena, Alin M. and Kov. J. Chem. Phys. , number =

    work page

  24. [25]

    Arslan Mazitov and Filippo Bigi and Matthias Kellner and Paolo Pegolo and Davide Tisi and Guillaume Fraux and Sergey Pozdnyakov and Philip Loche and Michele Ceriotti , journal =

    work page

  25. [26]

    Millan, Reisel and Bello-Jurado, Estefan\'

    work page

  26. [27]

    Reweighting-free-energies , url=

    work page

  27. [28]

    Andrew Pohorille and Christopher Jarzynski and Christophe Chipot , journal =

    work page

  28. [29]

    Albert Reuther and Jeremy Kepner and Chansup Byun and Siddharth Samsi and William Arcand and David Bestor and Bill Bergeron and Vijay Gadepally and Michael Houle and Matthew Hubbell and Michael Jones and Anna Klein and Lauren Milechin and Julia Mullen and Andrew Prout and Antonio Rosa and Charles Yee and Peter Michaleas , booktitle =

    work page

  29. [31]

    Andrea Rizzi and Paolo Carloni and Michele Parrinello , journal =

    work page

  30. [32]

    Wenhui Shen and Tong Zhou and Xinghua Shi , journal =

    work page

  31. [33]

    Michael R Shirts and John D Chodera , journal =

    work page

  32. [34]

    So Takamoto and Chikashi Shinagawa and Daisuke Motoki and Kosuke Nakago and Wenwen Li and Iori Kurata and Taku Watanabe and Yoshihiro Yayama and Hiroki Iriguchi and Yusuke Asano and others , journal =

    work page

  33. [35]

    Pratyush Tiwary and Axel van de Walle , journal =

    work page

  34. [36]

    Pauli Virtanen and Ralf Gommers and Evgeni Burovski and Travis E Oliphant and Warren Weckesser and David Cournapeau and Pearu Peterson and Tyler Reddy and Matt Haberland and Josh Wilson and others , journal =

    work page

  35. [37]

    Jia-Ning Wang and Wei Liu and Pengfei Li and Yan Mo and Wenxin Hu and Jun Zheng and Xiaoliang Pan and Yihan Shao and Ye Mei , journal =

    work page

  36. [38]

    Brandon M Wood and Misko Dzamba and Xiang Fu and Meng Gao and Muhammed Shuaibi and Luis Barroso-Luque and Kareem Abdelmaqsoud and Vahe Gharakhanyan and John R Kitchin and Daniel S Levine and others , journal =

    work page

  37. [39]

    Yi Isaac Yang and Qiang Shao and Jun Zhang and Lijiang Yang and Yi Qin Gao , journal =

    work page

  38. [40]

    Eric C-Y Yuan and Yunsheng Liu and Junmin Chen and Peichen Zhong and Sanjeev Raja and Tobias Kreiman and Santiago Vargas and Wenbin Xu and Martin Head-Gordon and Chao Yang and others , journal =

    work page

  39. [41]

    Phase-space overlap measures. II. Design and implementation of staging methods for free-energy calculations , author=. The Journal of chemical physics , volume=. 2005 , publisher=

    work page 2005

  40. [42]

    High-temperature equation of state by a perturbation method. I. Nonpolar gases , author=. The Journal of Chemical Physics , volume=. 1954 , publisher=

    work page 1954

  41. [43]

    Chemistry of Materials , volume=

    The free energy landscape of Li-and Na-ion transport in nanoconfinement with machine learning interatomic potentials , author=. Chemistry of Materials , volume=. 2026 , publisher=

    work page 2026

  42. [44]

    Li, Pengfei and Jia, Xiangyu and Pan, Xiaoliang and Shao, Yihan and Mei, Ye , journal =

    work page

  43. [45]

    ACS nano , volume=

    On the physical origins of reduced ionic conductivity in nanoconfined electrolytes , author=. ACS nano , volume=. 2025 , publisher=

    work page 2025

  44. [46]

    Understanding molecular simulation: from algorithms to applications , author=

    work page

  45. [47]

    doi:10.48550/arXiv.2206.07697 , title =

    Ilyes Batatia and D\'. doi:10.48550/arXiv.2206.07697 , title =

    work page doi:10.48550/arxiv.2206.07697

  46. [48]

    and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and

    Virtanen, Pauli and Gommers, Ralf and Oliphant, Travis E. and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and. Nature Methods , year =

    work page

  47. [50]

    Nature Computational Science , pages=

    High-throughput transition-state searches in zeolite nanopores , author=. Nature Computational Science , pages=. 2026 , publisher=

    work page 2026

  48. [51]

    Majumdar, Sauradeep and Steiner, Miguel and Dietschreit, Johannes C. B. and Roy, Swagata and Willimetz, Daniel and Grajciar, Lukáš and G\'omez-Bombarelli, Rafael , title =

    work page

  49. [52]

    , Smit , B

    botherref Frenkel , D. , Smit , B. : Understanding molecular simulation: from algorithms to applications (2023) botherref

    work page 2023

  50. [53]

    , Leli \`e vre , T

    botherref H \'e nin , J. , Leli \`e vre , T. , Shirts , M.R. , Valsson , O. , Delemotte , L. : Enhanced sampling methods for molecular dynamics simulations . arXiv preprint arXiv:2202.04164 (2022) botherref

    work page arXiv 2022

  51. [54]

    , Walle , A

    botherref Tiwary , P. , Walle , A. : A review of enhanced sampling approaches for accelerated molecular dynamics . Multiscale materials modeling for nanomechanics, 195--221 (2016) botherref

    work page 2016

  52. [55]

    , Shao , Q

    botherref Yang , Y.I. , Shao , Q. , Zhang , J. , Yang , L. , Gao , Y.Q. : Enhanced sampling in molecular dynamics . J. Chem. Phys. 151(7) (2019) botherref

    work page 2019

  53. [56]

    , Zhou , T

    barticle Shen , W. , Zhou , T. , Shi , X. : Enhanced sampling in molecular dynamics simulations and their latest applications -- A review . Nano Res. 16 ( 12 ), 13474 -- 13497 ( 2023 ) barticle

    work page 2023

  54. [57]

    , Chodera , J.D

    botherref Shirts , M.R. , Chodera , J.D. : Statistically optimal analysis of samples from multiple equilibrium states . J. Chem. Phys. 129(12) (2008) botherref

    work page 2008

  55. [58]

    , Rosenberg , J.M

    barticle Kumar , S. , Rosenberg , J.M. , Bouzida , D. , Swendsen , R.H. , Kollman , P.A. : The weighted histogram analysis method for free-energy calculations on biomolecules. I. The method . J. Comput. Chem. 13 ( 8 ), 1011 -- 1021 ( 1992 ) barticle

    work page 1992

  56. [59]

    , Pohorille , A

    bbook Chipot , C. , Pohorille , A. : Free Energy Calculations vol. 86 . Springer , ??? ( 2007 ) bbook

    work page 2007

  57. [60]

    , Diestler , D.J

    barticle Dietschreit , J.C. , Diestler , D.J. , G \'o mez-Bombarelli , R. : Entropy and Energy Profiles of Chemical Reactions . J. Chem. Theory Comput. 19 ( 16 ), 5369 -- 5379 ( 2023 ) barticle

    work page 2023

  58. [61]

    , Bello-Jurado , E

    botherref Millan , R. , Bello-Jurado , E. , Moliner , M. , Boronat , M. , G \'o mez-Bombarelli , R. : Effect of Framework Composition and NH 3 on the Diffusion of Cu + in Cu-CHA Catalysts Predicted by Machine-Learning Accelerated Molecular Dynamics 9(11), 2044--2056 botherref

    work page 2044

  59. [62]

    , Benner , P

    botherref Batatia , I. , Benner , P. , Chiang , Y. , Elena , A.M. , Kov \'a cs , D.P. , Riebesell , J. , Advincula , X.R. , Asta , M. , Avaylon , M. , Baldwin , W.J. , et al.: A foundation model for atomistic materials chemistry . J. Chem. Phys. 163(18) (2025) botherref

    work page 2025

  60. [63]

    , Dzamba , M

    botherref Wood , B.M. , Dzamba , M. , Fu , X. , Gao , M. , Shuaibi , M. , Barroso-Luque , L. , Abdelmaqsoud , K. , Gharakhanyan , V. , Kitchin , J.R. , Levine , D.S. , et al.: UMA: A Family of Universal Models for Atoms . arXiv preprint arXiv:2506.23971 (2025) botherref

    work page arXiv 2025

  61. [64]

    , Vandenhaute , S

    botherref Rhodes , B. , Vandenhaute , S. , S imkus , V. , Gin , J. , Godwin , J. , Duignan , T. , Neumann , M. : Orb-v3: atomistic simulation at scale . arXiv preprint arXiv:2504.06231 (2025) botherref

    work page arXiv 2025

  62. [65]

    , Bigi , F

    barticle Mazitov , A. , Bigi , F. , Kellner , M. , Pegolo , P. , Tisi , D. , Fraux , G. , Pozdnyakov , S. , Loche , P. , Ceriotti , M. : PET-MAD as a lightweight universal interatomic potential for advanced materials modeling . Nat. Commun. 16 ( 1 ), 10653 ( 2025 ) barticle

    work page 2025

  63. [66]

    , Liu , Y

    botherref Yuan , E.C.-Y. , Liu , Y. , Chen , J. , Zhong , P. , Raja , S. , Kreiman , T. , Vargas , S. , Xu , W. , Head-Gordon , M. , Yang , C. , et al.: Foundation models for atomistic simulation of chemistry and materials . Nat. Rev. Chem., 1--19 (2026) botherref

    work page 2026

  64. [67]

    , Roy , S

    barticle Majumdar , S. , Roy , S. , Jun , K. , Steiner , M. , G \'o mez-Bombarelli , R. : The free energy landscape of li-and na-ion transport in nanoconfinement with machine learning interatomic potentials . Chemistry of Materials 38 ( 8 ), 3971 -- 3981 ( 2026 ) barticle

    work page 2026

  65. [68]

    : High-temperature equation of state by a perturbation method

    barticle Zwanzig , R.W. : High-temperature equation of state by a perturbation method. i. nonpolar gases . The Journal of Chemical Physics 22 ( 8 ), 1420 -- 1426 ( 1954 ) barticle

    work page 1954

  66. [69]

    , Hudson , P.S

    barticle K\"onig , G. , Hudson , P.S. , Boresch , S. , Woodcock , H.L. : Multiscale free energy simulations: An efficient method for connecting classical MD simulations to QM or QM/MM free energies using non-Boltzmann Bennett reweighting schemes . J. Chem. Theory Comput. 10 ( 4 ), 1406 -- 1419 ( 2014 ) barticle

    work page 2014

  67. [70]

    , Woodcock , H.L

    barticle Hudson , P.S. , Woodcock , H.L. , Boresch , S. : Use of interaction energies in QM/MM free energy simulations . J. Chem. Theory Comput. 15 ( 8 ), 4632 -- 4645 ( 2019 ) barticle

    work page 2019

  68. [71]

    , Jarzynski , C

    barticle Pohorille , A. , Jarzynski , C. , Chipot , C. : Good practices in free-energy calculations . J. Phys. Chem. B 114 ( 32 ), 10235 -- 10253 ( 2010 ) barticle

    work page 2010

  69. [72]

    , Kofke , D.A

    botherref Wu , D. , Kofke , D.A. : Phase-space overlap measures. ii. design and implementation of staging methods for free-energy calculations. The Journal of chemical physics 123(8) (2005) botherref

    work page 2005

  70. [73]

    , Jia , X

    barticle Li , P. , Jia , X. , Pan , X. , Shao , Y. , Mei , Y. : Accelerated Computation of Free Energy Profile at ab Initio Quantum Mechanical/Molecular Mechanics Accuracy via a Semi-Empirical Reference Potential. I. Weighted Thermodynamics Perturbation . J. Chem. Theory Comput. 14 ( 11 ), 5583 -- 5596 ( 2018 ) barticle

    work page 2018

  71. [74]

    , Chen , A.A

    barticle Cai , G. , Chen , A.A. , Lin , S. , Lee , D.J. , Yu , K. , Holoubek , J. , Yin , Y. , Mu , A.U. , Meng , Y.S. , Liu , P. , : Unravelling Ultrafast Li Ion Transport in Functionalized Metal--Organic Framework-Based Battery Electrolytes . Nano Lett. 23 ( 15 ), 7062 -- 7069 ( 2023 ) barticle

    work page 2023

  72. [75]

    , Grey , C.P

    barticle Fong , K.D. , Grey , C.P. , Michaelides , A. : On the physical origins of reduced ionic conductivity in nanoconfined electrolytes . ACS nano 19 ( 13 ), 13191 -- 13201 ( 2025 ) barticle

    work page 2025

  73. [76]

    , Valencia , S

    barticle P\' e rez-Botella , E. , Valencia , S. , Rey , F. : Zeolites in Adsorption Processes: State of the Art and Future Prospects . Chem. Rev. 122 , 17647 -- 17695 ( 2022 ) barticle

    work page 2022

  74. [77]

    botherref MACE . Github. Accessed: June 2025. https://github.com/ACEsuit/mace botherref

    work page 2025

  75. [78]

    , Shinagawa , C

    barticle Takamoto , S. , Shinagawa , C. , Motoki , D. , Nakago , K. , Li , W. , Kurata , I. , Watanabe , T. , Yayama , Y. , Iriguchi , H. , Asano , Y. , : Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements . Nat. Commun. 13 ( 1 ), 2991 ( 2022 ) barticle

    work page 2022

  76. [79]

    , S \' i pka , M

    botherref Erlebach , A. , S \' i pka , M. , Saha , I. , Nachtigall , P. , Heard , C.J. , Grajciar , L. : A reactive neural network framework for water-loaded acidic zeolites . Nat. Commun. 15(1) (2024) botherref

    work page 2024

  77. [80]

    , Bornes , C

    barticle Lei , C. , Bornes , C. , Bengtsson , O. , Erlebach , A. , Slater , B. , Grajciar , L. , Heard , C.J. : A machine learning approach for dynamical modelling of Al distributions in zeolites via 23 Na / 27 Al solid-state NMR . Faraday Discuss. 255 , 46 -- 71 ( 2025 ) barticle

    work page 2025

  78. [81]

    , Hoffman , A.J

    botherref Ferri-Vicedo , P. , Hoffman , A.J. , Singhal , A. , G \'o mez-Bombarelli , R. : High-throughput transition-state searches in zeolite nanopores. Nature Computational Science, 1--13 (2026) botherref

    work page 2026

  79. [82]

    , Brabson , L.M

    botherref Sriram , A. , Brabson , L.M. , Yu , X. , Choi , S. , Abdelmaqsoud , K. , Moubarak , E. , Haan , P. , L \"o we , S. , Brehmer , J. , Kitchin , J.R. , et al.: The open dac 2025 dataset for sorbent discovery in direct air capture. arXiv preprint arXiv:2508.03162 (2025) botherref

    work page arXiv 2025

  80. [83]

    , Liu , R

    botherref Kaplan , A.D. , Liu , R. , Qi , J. , Ko , T.W. , Deng , B. , Riebesell , J. , Ceder , G. , Persson , K.A. , Ong , S.P. : A foundational potential energy surface dataset for materials . arXiv preprint arXiv:2503.04070 (2025) botherref

    work page arXiv 2025

Showing first 80 references.