Pith. sign in

REVIEW 3 major objections 7 minor 64 references

Reactive Chemistry at Unrestricted Coupled Cluster Level: High-throughput Calculations for Training Machine Learning Potentials

T0 review · 3 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Unrestricted CCSD(T) can be automated to supply a 3,119-structure reactive dataset, and potentials trained on it beat DFT-trained potentials in force and activation-energy accuracy.

desk verdict A genuinely useful UCCSD(T) reactive dataset and transferable MLIP, but the force basis-set correction is the load-bearing assumption and it is only validated near equilibrium. read the letter →

arxiv 2509.10872 v1 pith:KC6E5TTO submitted 2025-09-13 physics.chem-ph

Alice E. A. Allen , Rui Li , Sakib Matin , Xing Zhang , Benjamin Nebgen , Nicholas Lubbers , Justin S. Smith , Richard Messerly
show 3 more authors
Sergei Tretiak Garnet Kin-Lic Chan Kipton Barros
This is my paper · ORCID
classification physics.chem-ph
keywords unrestrictedcoupledclusterCCSD(T)machinelearninginteratomicpotentialactivebasissetcorrectionactivationenergytransferreactivechemistry
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that unrestricted CCSD(T)—the coupled-cluster method with singles, doubles, and perturbative triples—can be automated well enough to generate thousands of energies and forces for reactive organic molecules, and that machine-learned interatomic potentials trained on those labels are more accurate than potentials trained on DFT data. The authors build a 3,119-configuration gas-phase dataset of C/H/N/O molecules at a composite UCCSD(T)/QZ* level, using active learning, transition-state searches, and bond-stretching sampling. The concrete payoff is quantitative: switching the training data from DFT to UCCSD(T) improves force accuracy by more than 0.1 eV/Å and activation-energy reproduction by more than 0.1 eV, with the UCCSD(T)-trained potential reproducing held-out reaction barriers at 0.252 eV RMSE. A sympathetic reader would care because reactive chemistry—bond breaking, radicals, transition states—is exactly where DFT is known to be weakest, so a transferable coupled-cluster-level potential would make high-fidelity reaction simulation affordable.

What carries the argument

The load-bearing identity is the composite basis-set-corrected UCCSD(T) force, $$F_{\mathrm{UCCSD(T)/QZ^*}} = F_{\mathrm{UCCSD(T)/DZ}} + (F_{\mathrm{UCCSD/TZ}} - F_{\mathrm{UCCSD/DZ}}) + (F_{\mathrm{UMP2/QZ}} - F_{\mathrm{UMP2/TZ}})$$. It is what makes thousands of coupled-cluster force labels affordable, because it replaces a direct UCCSD(T)/QZ force evaluation with a DZ calculation plus two cheaper corrections, and the paper shows the corrected forces track explicit UCCSD(T)/QZ forces with a per-component RMSE of 0.073 eV/Å. The companion mechanism is an automated pipeline that selects a stable unrestricted Hartree-Fock reference with stability analysis and discards structures near Coulson-Fischer points, where the energy is continuous but the force is not; this filtering keeps inaccurate labels out of the training set.

What would settle it

Compute explicit UCCSD(T)/QZ forces for a sample of the dataset's bond-stretched and radical transition-state structures and compare them component-by-component with the UCCSD(T)/QZ* values; if the per-component RMSE grows well beyond the 0.073 eV/Å reported for small equilibrium molecules, or if the errors correlate with bond length or spin contamination, the basis-set correction—and therefore every MLIP label—is systematically biased.

Watch

Extended reading notes

Core claim

The central discovery is that the expensive UCCSD(T) level is a practical training source for reactive machine-learned interatomic potentials once three obstacles are automated: choosing a stable unrestricted Hartree-Fock reference by stability analysis, correcting basis-set incompleteness in both energies and forces with a composite scheme, and filtering structures close to Hartree-Fock instability where forces are undefined. With that workflow, the paper computes energies and forces for 3,119 gas-phase organic configurations and fine-tunes a neural-network interatomic potential to them. That potential reproduces held-out UCCSD(T) forces and activation energies more accurately than the same model trained on 270,720 DFT structures, and for activation energies it also beats the ωB97X DFT functional itself (0.252 eV versus 0.297 eV RMSE). The authors conclude that the level of theory of the training data, rather than the model family, is the current bottleneck limiting the accuracy of reactive machine-learned potentials.

Load-bearing premise

The load-bearing premise is that the basis-set incompleteness error in forces is transferable between MP2, UCCSD, and UCCSD(T), so the composite correction built from cheap DZ/TZ/QZ calculations faithfully reproduces UCCSD(T)/QZ forces; the paper validates this only on a few small near-equilibrium molecules, not on the stretched, radical, and transition-state geometries that dominate the reactive dataset.

Editorial extensions

If this is right

  • The UCCSD(T)-trained potential reproduces forces on transition states, reactants, and products with an RMSE more than 0.1 eV/Å lower than the DFT-trained potential.
  • Activation energies of reactions outside the training set are reproduced at 0.252 eV RMSE, better than the 0.297 eV RMSE of the ωB97X DFT functional and the 0.377 eV RMSE of the DFT-trained potential.
  • The UCCSD(T)-trained potential can drive NEB and dimer transition-state searches, converging to transition states and minimum-energy paths in minutes, where direct UCCSD(T)/QZ* searches would be prohibitively expensive.
  • Fine-tuning a pre-trained foundation model to the same 3,119-structure dataset yields a force error comparable to the transfer-learned model, suggesting the dataset itself—not the particular architecture—is what delivers the accuracy gain.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Extension: a direct audit of the composite basis-set correction on the hardest geometries in the dataset would be the cheapest stress test; computing true UCCSD(T)/QZ forces for a sample of stretched and radical structures and comparing them with UCCSD(T)/QZ* would show whether the 0.073 eV/Å benchmark holds where it matters most.
  • Extension: the paper's filtering of near-Coulson-Fischer structures delimits the potential's reliable domain to single-reference territory; automated multireference workflows, which the paper names as future work, would be the natural route to cover the excluded bond-breaking regions.
  • Extension: an ablation that removes the 1,953 bond-stretching structures from the training set could reveal whether the activation-energy advantage comes from transition-state labels or from stretched-geometry labels, telling future dataset builders which sampling earns its cost.
  • Extension: the two-stage recipe—cheap DFT exploration with active learning, then UCCSD(T) labels only on uncertain structures—should transfer to condensed-phase problems if cluster-extraction methods mature, which is the paper's stated open challenge.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. This manuscript presents an automated workflow for generating unrestricted CCSD(T) energies and forces for gas-phase organic reactions, using UHF stability analysis, a composite basis-set correction (Eqs. (1)–(2)), and a filtering protocol. The resulting dataset contains 3119 configurations drawn from transition states, reactants, products, NEB/dimer/SEGS sampling, and bond-stretched structures. Using transfer learning from a DFT-trained HIP-HOP-NN model, the authors fine-tune an MLIP on this UCCSD(T) dataset and report force RMSE improvements over a DFT-trained MLIP of more than 0.1 eV/Å and an activation-energy RMSE improvement from 0.377 eV to 0.252 eV. The paper also analyzes DFT vs UCCSD(T) differences across the dataset and fine-tunes MACE-MP as a second baseline architecture.

Significance. The significance is high if the central claim holds: a transferable reactive MLIP at CCSD(T) accuracy would be a substantial advance over DFT-trained potentials, and the dataset plus ALF code release would be a community resource. The paper has clear strengths: the workflow is described in detail; internal consistency checks include a 0.073 eV/Å validation of the force correction on small molecules; active learning and transfer learning are well motivated; activation-energy evaluations are performed on held-out reaction paths; and a second model family (MACE-MP) is fine-tuned as a comparison. The main risk is that the load-bearing basis-set correction is validated only on near-equilibrium six-atom molecules, with an error comparable to the claimed improvement, while the reactive geometries in the dataset are precisely where the correction's transferability is least tested. This concern does not invalidate the approach, but it must be addressed before the quantitative claims can be accepted.

major comments (3)
  1. [II.C.2 / SI S1.2, Eq. (2)] The central force-accuracy claim depends on Eq. (2), whose key assumption—that basis-set incompleteness errors in forces are transferable between MP2, UCCSD, and UCCSD(T)—is validated in SI S1.2 only on 300 K MD configurations of CO2, H2O, C2H2, and H2CO. These are near-equilibrium six-atom species, and the reported correction error of 0.073 eV/Å per force component is the same order of magnitude as the >0.1 eV/Å improvement claimed in the abstract. Since the dataset deliberately includes 1953 bond-stretched structures and NEB/dimer/SEGS geometries with force components up to tens of eV/Å (Fig. S11), the transferability of the correction at stretched, radical, and transition-state geometries is a load-bearing assumption. Please add validation of Eq. (2) on representative reactive geometries, or provide a bound on the correction error for those geometries.
  2. [SI S1.3 / S1.4] The filtering protocol removes structures near the Coulson–Fischer point and structures with large spin-state differences across bases, but it does not establish the accuracy of Eq. (2) on the structures that remain. The manuscript itself documents large basis-set dependence of the UHF reference for a removed transition state (⟨S2⟩ = 0.51/0.22/0.13 for DZ/TZ/QZ and an orbital Hessian eigenvalue below 10^-3 Ha), and the filter thresholds of 1 eV/Å and 5 eV/Å are much larger than the claimed 0.1 eV/Å force improvement. A systematic failure of the correction on retained reactive structures would propagate into all 3119 force labels and could create or mask the reported MLIP improvement. Please quantify how the final training labels and the MLIP comparison depend on these thresholds and on the correction assumptions.
  3. [II.C.2, Eq. (1)] The same transferability assumption underlies the energy correction in Eq. (1), and the activation-energy results in Figs. 5 and 7 are based on these composite energies. No validation of Eq. (1) at transition-state geometries is provided; the SI energy validation shown in Fig. S3 is limited to the same near-equilibrium small molecules. Please either validate Eq. (1) on representative transition-state and stretched geometries or state the expected uncertainty in the activation-energy RMSE values.
minor comments (7)
  1. [Title / throughout] The title contains a line-break typo, "high-throug hput"; please correct it.
  2. [SI Fig. S2] The caption reports the RMSE as "0.073 eV/" with an incomplete unit; please specify eV/Å per force component.
  3. [SI S1.2] The text refers to "UCCD(T)/QZ" in one passage; this should be "UCCSD(T)/QZ*" to avoid ambiguity.
  4. [SI S1.3.1 and S1.4] The SI figure numbering is inconsistent: both S1.3.1 and S1.4 reference "Fig. S10" for different content; please renumber the figures and update all references.
  5. [Fig. 6 / Section III.B] The main text does not give the numeric test-set force RMSE values for the DFT-trained and UCCSD(T)-trained HIP-HOP models; please include these values in the caption or text so the claimed >0.1 eV/Å improvement can be verified directly.
  6. [Fig. 7 / Section III.B] The activation-energy numbers are reported inconsistently: the text states 0.377 eV RMSE for the DFT-trained MLIP, while the Fig. 7 caption reports 0.297 eV for the ωB97X functional; please clarify which comparison supports the abstract statement of "over 0.1 eV" improvement.
  7. [SI S6.1 / Fig. S13] In the bond-dissociation discussion the DFT reference is described only as "singlet DFT"; please specify the exact functional and basis set used in Fig. S13.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the MLIP is trained on UCCSD(T) labels and evaluated on held-out reaction paths; the basis-set correction is an explicit, independently validated approximation rather than a fitted prediction.

full rationale

The paper's central derivation chain is self-contained and out-of-sample. UCCSD(T)/QZ* labels are constructed by the explicit composite correction of Eqs. (1) and (2), where each term is a separate quantum-chemical calculation; no parameter is fitted to the MLIP test set, and no predicted quantity is defined from the training target. The correction's additivity assumption is validated in SI S1.2 against direct UCCSD(T)/QZ forces on small molecules (RMSE 0.073 eV/A), so any failure at stretched or radical geometries is a correctness/transferability risk, not a circular step. The MLIP is trained on a subset of UCCSD(T) data and then evaluated on reaction paths explicitly stated to be excluded from the UCCSD(T) training set, giving genuine out-of-sample force and activation-energy predictions. The DFT-trained and UCCSD(T)-trained potentials are compared against the same UCCSD(T) reference, which is a legitimate benchmark rather than a construction-equals-prediction artifact. Self-citations (HIP-NN, ALF, HIP-HOP-NN) supply software and model architectures used as tools; the paper's conclusions do not rest on a self-citation chain or an imported uniqueness theorem. No equation reduces to its inputs, and no fitted parameter is renamed as a prediction.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

No new particles, forces, or conserved quantities are introduced. UCCSD(T)/QZ* is a composite computational model, not a new physical entity. The load-bearing assumptions are the transferability of the basis set correction for forces (Eq. 2) and the sufficiency of single-reference UCCSD(T) after filtering. The only free parameter is the set of hand-chosen filtering thresholds.

free parameters (1)
  • UCCSD(T) filtering thresholds = 1 eV/Å (R vs U force difference), 1 eV/Å (MP2 force correction), 5 eV/Å (UCCSD force correction), 0.1 (S2 difference)
    Chosen by hand to remove structures near Hartree-Fock instabilities; these thresholds determine which configurations are excluded and thus shape the DFT versus UCCSD(T) comparison and the MLIP training set.
assumptions (5)
  • domain assumption UCCSD(T) is the reference of truth for reactive energies and forces.
    The dataset and all benchmarks treat UCCSD(T)/QZ* as gold standard; no multi-reference validation is performed for the filtered structures.
  • ad hoc to paper Basis set incompleteness error in forces is transferable across UMP2, UCCSD, and UCCSD(T).
    Eq. (2) in Methods II.C.2; validated only on six-atom molecules near equilibrium in SI S1.2.
  • domain assumption Restricted Kohn-Sham singlet DFT is an acceptable reference for the reactive DFT dataset.
    All DFT data use RKS singlet states; open-shell regions are only covered by the UCCSD(T) calculations, limiting the DFT comparison.
  • domain assumption The HIP-HOP-NN architecture is expressive enough to represent the reactive potential energy surface.
    The claim that the MLIP can replace DFT depends on model capacity; the paper uses an ensemble of 8 models but does not prove convergence.
  • domain assumption Hartree-Fock stability analysis and S2 filtering ensure the UCCSD(T) results remain single-reference.
    Used in Methods II.C.2 and SI S1.4; the filtering criteria are thresholds, not a rigorous error bound.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reactive Chemistry at Unrestricted Coupled Cluster Level: High-throughput Calculations for Training Machine Learning Potentials." pith.science (2026). https://pith.science/paper/KC6E5TTO

@misc{pith2026250910872,
  author       = {Pith},
  title        = {Pith review of: Reactive Chemistry at Unrestricted Coupled Cluster Level: High-throughput Calculations for Training Machine Learning Potentials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KC6E5TTO}},
  note         = {Machine review of arXiv:2509.10872}
}
read the original abstract

Accurately modeling chemical reactions at the atomistic level requires high-level electronic structure theory due to the presence of unpaired electrons and the need to properly describe bond breaking and making energetics. Commonly used approaches such as Density Functional Theory (DFT) frequently fail for this task due to deficiencies that are well recognized. However, for high-fidelity approaches, creating large datasets of energies and forces for reactive processes to train machine learning interatomic potentials or force fields is daunting. For example, the use of the unrestricted coupled cluster level of theory has previously been seen as unfeasible due to high computational costs, the lack of analytical gradients in many computational codes, and additional challenges such as constructing suitable basis set corrections for forces. In this work, we develop new methods and workflows to overcome the challenges inherent to automating unrestricted coupled cluster calculations. Using these advancements, we create a dataset of gas-phase reactions containing energies and forces for 3119 different organic molecules configurations calculated at the gold-standard level of unrestricted CCSD(T) (coupled cluster singles doubles and perturbative triples). With this dataset, we provide an analysis of the differences between the density functional and unrestricted CCSD(T) descriptions. We develop a transferable machine learning interatomic potential for gas-phase reactions, trained on unrestricted CCSD(T) data, and demonstrate the advantages of transitioning away from DFT data. Transitioning from training to DFT to training to UCCSD(T) datasets yields an improvement of more than 0.1 eV/{\AA} in force accuracy and over 0.1 eV in activation energy reproduction.

Figures

Figures reproduced from arXiv: 2509.10872 by the authors.

Figure 1
Figure 1. Unrestricted CCSD(T) calculations sampled with a varie [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The forces for UCCSD(T)/DZ compared to the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. The reactants, products and transition states for [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: A comparison of forces for DFT functionals calculated in th [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The activation energy for DFT functionals in comparison t [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: A comparison of forces calculated with the machine learn [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: The activation energies for the HIP-HOP-NN model [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: The transition states with the MLIP trained to [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 46 canonical work pages

  1. [1]

    author author W. L. \ Jorgensen \ and\ author J. Tirado-Rives ,\ @noop journal journal J. Am. Chem. Soc. \ volume 110 ,\ pages 1657 ( year 1988 ) NoStop

    work page 1988

  2. [2]

    author author S. J. \ Weiner , author P. A. \ Kollman , author D. A. \ Case , author U. C. \ Singh , author C. Ghio , author G. Alagona , author S. Profeta , \ and\ author P. Weiner ,\ @noop journal journal J. Am. Chem. Soc. \ volume 106 ,\ pages 765 ( year 1984 ) NoStop

    work page 1984

  3. [3]

    author author J. S. \ Smith , author B. T. \ Nebgen , author R. Zubatyuk , author N. Lubbers , author C. Devereux , author K. Barros , author S. Tretiak , author O. Isayev , \ and\ author A. E. \ Roitberg ,\ @noop journal journal Nat. Commun. \ volume 10 ,\ pages 2903 ( year 2019 ) NoStop

    work page 2019

  4. [4]

    author author J. S. \ Smith , author O. Isayev , \ and\ author A. E. \ Roitberg ,\ @noop journal journal Chem. Sci. \ volume 8 ,\ pages 3192 ( year 2017 ) NoStop

    2017

  5. [5]

    Chmiela , author K.-R

    author author S. Chmiela , author K.-R. \ Sauceda , Huziel E.and M \"u ller , \ and\ author A. Tkatchenko ,\ @noop journal journal Nat. Commun. \ volume 9 ,\ pages 3887 ( year 2018 ) NoStop

    work page 2018

  6. [6]

    Zhao , author O

    author author Y. Zhao , author O. Tishchenko , author J. R. \ Gour , author W. Li , author J. J. \ Lutz , author P. Piecuch , \ and\ author D. G. \ Truhlar ,\ 10.1021/jp811054n journal journal J. Phys. Chem. A \ volume 113 ,\ pages 5786 ( year 2009 ) ,\ note pMID: 19374412 NoStop

    work page doi:10.1021/jp811054n

  7. [7]

    Goerigk \ and\ author S

    author author L. Goerigk \ and\ author S. Grimme ,\ 10.1021/ct900489g journal journal J. Chem. Theory Comput. \ volume 6 ,\ pages 107 ( year 2010 ) NoStop

    doi:10.1021/ct900489g 2010

  8. [8]

    Zhao \ and\ author D

    author author Y. Zhao \ and\ author D. G. \ Truhlar ,\ 10.1021/ct1006604 journal journal J. Chem. Theory Comput. \ volume 7 ,\ pages 669 ( year 2011 ) NoStop

    work page doi:10.1021/ct1006604 2011

Show all 64 references
  1. [9]

    author author E. R. \ Johnson , author P. Mori-Sánchez , author A. J. \ Cohen , \ and\ author W. Yang ,\ 10.1063/1.3021474 journal journal J. Chem. Phys. \ volume 129 ,\ pages 204112 ( year 2008 ) NoStop

    2008 doi

  2. [10]

    author author A. A. \ Chamkin \ and\ author E. S. \ Serkova ,\ https://doi.org/10.1002/jcc.26398 journal journal J. Comput. Chem. \ volume 41 ,\ pages 2388 ( year 2020 ) NoStop

    2020 doi

  3. [11]

    Vermeeren , author M

    author author P. Vermeeren , author M. Dalla Tiezza , author M. E. \ Wolf , author M. E. \ Lahm , author W. D. \ Allen , author H. F. \ Schaefer , author T. A. \ Hamlin , \ and\ author F. M. \ Bickelhaupt ,\ 10.1039/D2CP02234F journal journal Phys. Chem. Chem. Phys. \ volume 2...

    doi

  4. [12]

    author author A. J. \ Cohen , author P. Mori-Sánchez , \ and\ author W. Yang ,\ 10.1126/science.1158722 journal journal Science \ volume 321 ,\ pages 792 ( year 2008 ) NoStop

    2008 doi

  5. [13]

    Li , author M

    author author H. Li , author M. Mansoori Kermani , author A. Ottochian , author O. Crescenzi , author B. G. \ Janesko , author D. G. \ Truhlar , author G. Scalmani , author M. J. \ Frisch , author I. Ciofini , \ and\ author C. Adamo ,\ 10.1021/jacs.3c12713 journal journal J. A...

    doi

  6. [14]

    u tt , author F. Arbabzadah , author S. Chmiela , author K.-R. \ M \

    author author K. T. \ Sch \"u tt , author F. Arbabzadah , author S. Chmiela , author K.-R. \ M \"u ller , \ and\ author A. Tkatchenko ,\ @noop journal journal Nat. Commun. \ volume 8 ,\ pages 13890 ( year 2017 ) NoStop

    2017

  7. [15]

    author author A. P. \ Bart \'o k , author M. C. \ Payne , author R. Kondor , \ and\ author G. Cs \'a nyi ,\ @noop journal journal Phys. Rev. Lett. \ volume 104 ,\ pages 136403 ( year 2010 ) NoStop

    2010

  8. [16]

    Behler ,\ @noop journal journal J

    author author J. Behler ,\ @noop journal journal J. Chem. Phys \ volume 134 ,\ pages 074106 ( year 2011 ) NoStop

    2011

  9. [17]

    u tt , author A. Tkatchenko , \ and\ author K.-R. \ M \

    author author O. T. \ Unke , author S. Chmiela , author H. E. \ Sauceda , author M. Gastegger , author I. Poltavsky , author K. T. \ Sch \"u tt , author A. Tkatchenko , \ and\ author K.-R. \ M \"u ller ,\ @noop journal journal Chem. Rev. \ volume 121 ,\ pages 10142 ( year 2021...

    2021

  10. [18]

    Mueller , author A

    author author T. Mueller , author A. Hernandez , \ and\ author C. Wang ,\ @noop journal journal J. Chem. Phys. \ volume 152 ,\ pages 050902 ( year 2020 ) NoStop

    2020

  11. [19]

    author author A. P. \ Bart \'o k , author J. Kermode , author N. Bernstein , \ and\ author G. Cs \'a nyi ,\ @noop journal journal Phys. Rev. X \ volume 8 ,\ pages 041048 ( year 2018 ) NoStop

    2018

  12. [20]

    author author R. J. \ Bartlett \ and\ author M. Musia ,\ 10.1103/RevModPhys.79.291 journal journal Rev. Mod. Phys. \ volume 79 ,\ pages 291 ( year 2007 ) NoStop

    2007 doi

  13. [21]

    Visscher , author T

    author author L. Visscher , author T. J. \ Lee , \ and\ author K. G. \ Dyall ,\ 10.1063/1.472655 journal journal J. Chem. Phys. \ volume 105 ,\ pages 8769 ( year 1996 ) NoStop

    1996 doi

  14. [22]

    Korth \ and\ author S

    author author M. Korth \ and\ author S. Grimme ,\ 10.1021/ct800511q journal journal J. Chem. Theory Comput. \ volume 5 ,\ pages 993 ( year 2009 ) NoStop

    2009 doi

  15. [23]

    Hait \ and\ author M

    author author D. Hait \ and\ author M. Head-Gordon ,\ @noop journal journal J. Phys. Chem. Letters \ volume 9 ,\ pages 6280 ( year 2018 ) NoStop

    2018

  16. [24]

    Seeger \ and\ author J

    author author R. Seeger \ and\ author J. A. \ Pople ,\ @noop journal journal J. Chem. Phys. \ volume 66 ,\ pages 3045 ( year 1977 ) NoStop

    1977

  17. [25]

    Fan , author J

    author author Y. Fan , author J. Ho , \ and\ author R. P. \ Bettens ,\ @noop journal journal J. Phys. Chem. A \ volume 110 ,\ pages 2796 ( year 2006 ) NoStop

    2006

  18. [26]

    author author J. S. \ Smith , author B. Nebgen , author N. Lubbers , author O. Isayev , \ and\ author A. E. \ Roitberg ,\ @noop journal journal J. Chem. Phys. \ volume 148 ,\ pages 241733 ( year 2018 ) NoStop

    2018

  19. [27]

    author author C. A. \ Grambow , author L. Pattanaik , \ and\ author W. H. \ Green ,\ 10.1038/s41597-020-0460-4 journal journal Sci. Data \ volume 7 ,\ pages 137 ( year 2020 ) NoStop

    2020 doi

  20. [28]

    Henkelman \ and\ author H

    author author G. Henkelman \ and\ author H. Jónsson ,\ @noop journal journal J. Chem. Phys. \ volume 111 ,\ pages 7010 ( year 1999 ) NoStop

    1999

  21. [29]

    Jonsson , author G

    author author H. Jonsson , author G. Mills , \ and\ author K. W. \ Jacobsen ,\ 10.1142/9789812839664_0016 title Classical and Quantum Dynamics in Condensed Phase \ ( year 1998 )\ pp.\ pages 385--404 NoStop

    1998 doi

  22. [30]

    author author P. M. \ Zimmerman ,\ https://doi.org/10.1002/jcc.23833 journal journal J. Comput. Chem. \ volume 36 ,\ pages 601 ( year 2015 ) NoStop

    2015 doi

  23. [31]

    author author D. P. \ Kovács , author J. H. \ Moore , author N. J. \ Browning , author I. Batatia , author J. T. \ Horton , author Y. Pu , author V. Kapil , author W. C. \ Witt , author I.-B. \ Magdău , author D. J. \ Cole , \ and\ author G. Csányi ,\ 10.1021/jacs.4c07099 jour...

    doi

  24. [32]

    Zubatyuk , author J

    author author R. Zubatyuk , author J. S. \ Smith , author J. Leszczynski , \ and\ author O. Isayev ,\ 10.1126/sciadv.aav6490 journal journal Science Advances \ volume 5 ,\ pages eaav6490 ( year 2019 ) NoStop

    2019 doi

  25. [33]

    Zhang , author M

    author author S. Zhang , author M. Z. \ Mako \' s , author R. B. \ Jadrich , author E. Kraka , author K. Barros , author B. T. \ Nebgen , author S. Tretiak , author O. Isayev , author N. Lubbers , author R. A. \ Messerly , \ and\ author J. S. \ Smith ,\ @noop journal journal N...

    2024

  26. [34]

    Schreiner , author A

    author author M. Schreiner , author A. Bhowmik , author T. Vegge , author J. Busk , \ and\ author O. Winther ,\ @noop journal journal Sci. Data \ volume 9 ,\ pages 779 ( year 2022 ) NoStop

    2022

  27. [35]

    Batatia , author P

    author author I. Batatia , author P. Benner , author Y. Chiang , author A. M. \ Elena , author D. P. \ Kovács , author J. Riebesell , author X. R. \ Advincula , author M. Asta , author M. Avaylon , author W. J. \ Baldwin , author F. Berger , author N. Bernstein , author A. Bho...

    2024 arXiv

  28. [36]

    Jain , author S

    author author A. Jain , author S. P. \ Ong , author G. Hautier , author W. Chen , author W. D. \ Richards , author S. Dacek , author S. Cholia , author D. Gunter , author D. Skinner , author G. Ceder , \ and\ author K. A. \ Persson ,\ 10.1063/1.4812323 journal journal APL Mate...

    doi

  29. [37]

    author author B. J. \ Braams \ and\ author J. M. \ Bowman ,\ @noop journal journal Int. Rev. Phys. Chem. \ volume 28 ,\ pages 577 ( year 2009 ) NoStop

    2009

  30. [38]

    Qu , author Q

    author author C. Qu , author Q. Yu , \ and\ author J. M. \ Bowman ,\ @noop journal journal Annu. Rev. Phys. Chem. \ volume 69 ,\ pages 151 ( year 2018 ) NoStop

    2018

  31. [39]

    Huang , author B

    author author X. Huang , author B. J. \ Braams , \ and\ author J. M. \ Bowman ,\ @noop journal journal J. Chem. Phys \ volume 122 ,\ pages 044308 ( year 2005 ) NoStop

    2005

  32. [40]

    author author Q. H. \ Hu , author A. M. \ Johannesen , author D. S. \ Graham , \ and\ author J. D. \ Goodpaster ,\ @noop journal journal Digit. Discov. \ volume 2 ,\ pages 1058 ( year 2023 ) NoStop

    2023

  33. [41]

    Chigaev , author J

    author author M. Chigaev , author J. S. \ Smith , author S. Anaya , author B. Nebgen , author M. Bettencourt , author K. Barros , \ and\ author N. Lubbers ,\ 10.1063/5.0142127 journal journal J. Chem. Phys. \ volume 158 ,\ pages 184108 ( year 2023 ) NoStop

    doi

  34. [42]

    Lubbers , author J

    author author N. Lubbers , author J. S. \ Smith , \ and\ author K. Barros ,\ @noop journal journal J. Chem. Phys. \ volume 148 ( year 2018 ) NoStop

    2018

  35. [43]

    author author A. E. A. \ Allen , author E. Shinkle , author R. Bujack , \ and\ author N. Lubbers ,\ https://arxiv.org/abs/2503.23515 title Optimal invariant bases for atomistic machine learning , \ ( year 2025 ),\ http://arxiv.org/abs/2503.23515 arXiv:2503.23515 [physics.chem-...

    2025 arXiv

  36. [44]

    author author S. J. \ Pan \ and\ author Q. Yang ,\ @noop journal journal IEEE Trans. Knowl. Data Eng. \ volume 22 ,\ pages 1345 ( year 2010 ) NoStop

    2010

  37. [45]

    Ramakrishnan , author P

    author author R. Ramakrishnan , author P. O. \ Dral , author M. Rupp , \ and\ author O. A. \ von Lilienfeld ,\ @noop journal journal Sci. Data \ volume 1 ,\ pages 140022 ( year 2014 ) NoStop

    2014

  38. [46]

    \ Chai \ and\ author M

    author author J.-D. \ Chai \ and\ author M. Head-Gordon ,\ 10.1063/1.2834918 journal journal J. Chem. Phys. \ volume 128 ,\ pages 084106 ( year 2008 ) NoStop

    2008 doi

  39. [47]

    Adamo \ and\ author V

    author author C. Adamo \ and\ author V. Barone ,\ 10.1063/1.478522 journal journal J. Chem. Phys. \ volume 110 ,\ pages 6158 ( year 1999 ) NoStop

    1999 doi

  40. [48]

    author author A. D. \ Becke ,\ @noop journal journal J. Chem. Phys. \ volume 98 ,\ pages 5648 ( year 1993 ) NoStop

    1993

  41. [49]

    author author P. J. \ Stephens , author F. J. \ Devlin , author C. F. \ Chabalowski , \ and\ author M. J. \ Frisch ,\ @noop journal journal J. Phys. Chem. \ volume 98 ,\ pages 11623 ( year 1994 ) NoStop

    1994

  42. [50]

    Neese , author F

    author author F. Neese , author F. Wennmohs , author U. Becker , \ and\ author C. Riplinger ,\ @noop journal journal J. Chem. Phys. \ volume 152 ( year 2020 ) NoStop

    2020

  43. [51]

    Sun , author X

    author author Q. Sun , author X. Zhang , author S. Banerjee , author P. Bao , author M. Barbry , author N. S. \ Blunt , author N. A. \ Bogdanov , author G. H. \ Booth , author J. Chen , author Z.-H. \ Cui , author J. J. \ Eriksen , author Y. Gao , author S. Guo , author J. Her...

    2020

  44. [52]

    Sun , author T

    author author Q. Sun , author T. C. \ Berkelbach , author N. S. \ Blunt , author G. H. \ Booth , author S. Guo , author Z. Li , author J. Liu , author J. D. \ McClain , author E. R. \ Sayfutyarova , author S. Sharma , author S. Wouters , \ and\ author G. K.-L. \ Chan ,\ @noop ...

    2018

  45. [53]

    author author J. T. \ Margraf , author A. Perera , author J. J. \ Lutz , \ and\ author R. J. \ Bartlett ,\ 10.1063/1.5003128 journal journal J. Chem. Phys. \ volume 147 ,\ pages 184101 ( year 2017 ) NoStop

    2017 doi

  46. [54]

    author author H. B. \ Schlegel ,\ @noop journal journal Encyclopedia of Computational Chemistry \ ( year 2002 ) NoStop

    2002

  47. [55]

    Coulson \ and\ author I

    author author C. Coulson \ and\ author I. Fischer ,\ @noop journal journal Lond. Edinb. Dubl. Phil. Mag. \ volume 40 ,\ pages 386 ( year 1949 ) NoStop

    1949

  48. [56]

    author author V. F. \ Grizzi , author N. Fedik , author S. Anaya , author J. Gomes , author Y. W. \ Li , author K. Barros , author N. Lubbers , author R. Messerly , author J. Smith , author B. Nebgen , \ and\ author S. Tretiak ,\ @noop title ALF , an active learning framework ...

  49. [57]

    Spiekermann , author L

    author author K. Spiekermann , author L. Pattanaik , \ and\ author W. H. \ Green ,\ @noop journal journal Scient. Data \ volume 9 ,\ pages 417 ( year 2022 ) NoStop

    2022

  50. [58]

    Chen , author P

    author author S. Chen , author P. Yu , \ and\ author K. N. \ Houk ,\ @noop journal journal J. Am. Chem. Soc. \ volume 140 ,\ pages 18124 ( year 2018 ) NoStop

    2018

  51. [59]

    author author J. J. \ Wardzala , author D. S. \ King , author L. Ogunfowora , author B. Savoie , \ and\ author L. Gagliardi ,\ 10.1021/acscentsci.3c01559 journal journal ACS Central Science \ volume 10 ,\ pages 833 ( year 2024 ) NoStop

    doi

  52. [60]

    author author D. S. \ King , author M. R. \ Hermes , author D. G. \ Truhlar , \ and\ author L. Gagliardi ,\ 10.1021/acs.jctc.2c00630 journal journal J. Chem. Theory Comput. \ volume 18 ,\ pages 6065 ( year 2022 ) ,\ note pMID: 36112354 NoStop

    2022 doi

  53. [61]

    Zhang , author C

    author author X. Zhang , author C. Li , author H.-Z. \ Ye , author T. C. \ Berkelbach , \ and\ author G. K.-L. \ Chan ,\ @noop journal journal J. Chem. Phys. \ volume 161 ,\ pages 014109 ( year 2024 b ) NoStop

    2024

  54. [62]

    Herbold \ and\ author J

    author author M. Herbold \ and\ author J. Behler ,\ @noop journal journal Phys. Chem. Chem. Phys. \ volume 25 ,\ pages 12979 ( year 2023 ) NoStop

    2023

  55. [63]

    S.; Bogdanov, N

    Sun, Q.; Zhang, X.; Banerjee, S.; Bao, P.; Barbry, M.; Blunt, N. S.; Bogdanov, N. A.; Booth, G. H.; Chen, J.; Cui, Z.-H.; Eriksen, J. J.; Gao, Y.; Guo, S.; Hermann, J.; Hermes, M. R.; Koh, K.; Koval, P.; Lehtola, S.; Li, Z.; Liu, J.; Mardirossian, N.; McClain, J. D.; Motta, M....

    2020

  56. [64]

    ]" " v8 KX D f O1| Hy@ 5

    Sun, Q.; Berkelbach, T. C.; Blunt, N. S.; Booth, G. H.; Guo, S.; Li, Z.; Liu, J.; McClain, J. D.; Sayfutyarova, E. R.; Sharma, S.; Wouters, S.; Chan, G. K.-L. PySCF: the Python-based simulations of chemistry framework. WIREs Comput. Mol. Sci. 2018, 8, e1340 mcitethebibliograph...

    2018

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.