pith. machine review for the scientific record. sign in

arxiv: 2604.12140 · v1 · submitted 2026-04-13 · 💻 cs.LG · cond-mat.mtrl-sci· physics.chem-ph

Recognition: unknown

XANE(3): An E(3)-Equivariant Graph Neural Network for Accurate Prediction of XANES Spectra from Atomic Structures

Vitor F. Grizzi , Luke N. Pretzie , Jiayi Xu , Cong Liu
Authors on Pith no claims yet

Pith reviewed 2026-05-10 14:55 UTC · model grok-4.3

classification 💻 cs.LG cond-mat.mtrl-sciphysics.chem-ph
keywords XANES spectraequivariant graph neural networkmachine learning for spectroscopyiron oxide materialsspectral predictionE(3) symmetryFDMNES simulations
0
0 comments X

The pith

An E(3)-equivariant graph neural network predicts XANES spectra directly from atomic structures.

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

The paper develops XANE(3), an equivariant graph neural network that forecasts X-ray absorption near-edge structure spectra straight from the positions and types of atoms. It uses symmetry-preserving message passing, attention centered on the absorbing atom, and training that matches not only the spectrum but also its first and second derivatives to preserve peak shapes. On over five thousand simulated spectra of iron oxide surface facets, the model reaches a mean squared error of 0.001 and reproduces main edges, pre-edge features, and post-edge oscillations. This approach matters because full XANES calculations remain expensive, so an accurate fast surrogate could speed up the interpretation of material electronic structure from spectroscopy data.

Core claim

XANE(3) is a physics-based E(3)-equivariant graph neural network that predicts XANES spectra from atomic structures by combining tensor-product message passing with spherical harmonic edge features, absorber-query attention pooling, custom equivariant layer normalization, adaptive gated residual connections, and a multi-scale Gaussian basis readout with optional sigmoidal background, trained with a composite objective of pointwise reconstruction plus first- and second-derivative matching, and it achieves a spectrum mean squared error of 1.0 × 10^{-3} on a test set of 5,941 FDMNES simulations of iron oxide surface facets while reproducing the main edge structure, relative peak intensities,pre

What carries the argument

The XANE(3) model architecture, which applies E(3)-equivariant tensor-product message passing and absorber-conditioned attention to atomic graphs to produce spectral outputs while preserving 3D symmetries.

Load-bearing premise

The assumption that accurate performance on FDMNES simulations of iron oxide facets will hold for experimental XANES spectra or for materials outside this narrow chemical and structural domain.

What would settle it

A direct comparison of model predictions against measured experimental XANES spectra for the same iron oxide atomic structures would test whether the reported accuracy generalizes beyond simulations.

Figures

Figures reproduced from arXiv: 2604.12140 by Cong Liu, Jiayi Xu, Luke N. Pretzie, Vitor F. Grizzi.

Figure 1
Figure 1. Figure 1: Schematic of XANE(3). The model maps an input atomic structure to a predicted [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Comparison between predicted and computed XANES spectra for representative test [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Examples showing that lower pointwise spectrum MSE does not necessarily imply better [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Comparison between predicted and computed XANES for the version of the model [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
read the original abstract

We present XANE(3), a physics-based E(3)-equivariant graph neural network for predicting X-ray absorption near-edge structure (XANES) spectra directly from atomic structures. The model combines tensor-product message passing with spherical harmonic edge features, absorber-query attention pooling, custom equivariant layer normalization, adaptive gated residual connections, and a spectral readout based on a multi-scale Gaussian basis with an optional sigmoidal background term. To improve line-shape fidelity, training is performed with a composite objective that includes pointwise spectral reconstruction together with first- and second-derivative matching terms. We evaluate the model on a dataset of 5,941 FDMNES simulations of iron oxide surface facets and obtain a spectrum mean squared error of $1.0 \times 10^{-3}$ on the test set. The model accurately reproduces the main edge structure, relative peak intensities, pre-edge features, and post-edge oscillations. Ablation studies show that the derivative-aware objective, custom equivariant normalization, absorber-conditioned attention pooling, adaptive gated residual mixing, and global background term each improve performance. Interestingly, a capacity-matched scalar-only variant achieves comparable pointwise reconstruction error but reduced derivative-level fidelity, indicating that explicit tensorial channels are not strictly required for low intensity error on this dataset, although they remain beneficial for capturing finer spectral structure. These results establish XANE(3) as an accurate and efficient surrogate for XANES simulation and offer a promising route toward accelerated spectral prediction, ML-assisted spectroscopy, and data-driven materials discovery.

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 / 1 minor

Summary. The paper introduces XANE(3), an E(3)-equivariant graph neural network for predicting XANES spectra from atomic structures. It employs tensor-product message passing with spherical harmonic features, absorber-query attention, custom equivariant normalization, adaptive gated residuals, and a multi-scale Gaussian readout with optional background term. Training uses a composite loss with pointwise, first-, and second-derivative terms. On a dataset of 5,941 FDMNES simulations of iron oxide surface facets, the model achieves a test MSE of 1.0 × 10^{-3}, accurately reproducing main edge, pre-edge, and post-edge features. Ablations confirm the value of each component, and a scalar-only variant is shown to match pointwise accuracy but lag in derivative fidelity.

Significance. If validated more broadly, this architecture could serve as an efficient surrogate for expensive XANES simulations, facilitating faster materials screening and spectroscopy analysis. The work is strengthened by detailed ablations and the inclusion of derivative matching in the objective, which improves line-shape fidelity. The observation regarding the scalar variant provides useful insight into the necessity of equivariant features for this task.

major comments (1)
  1. The reported results and the claim of an 'accurate and efficient surrogate for XANES simulation' are based entirely on performance within the narrow domain of FDMNES-simulated iron oxide surface facets. No evaluation on experimental XANES spectra or on materials outside iron oxides is provided. This untested sim-to-real and cross-chemistry transfer is load-bearing for the broader utility claimed in the abstract's final sentence, as failure to generalize would limit the model's applicability to ML-assisted spectroscopy.
minor comments (1)
  1. The abstract notes that the scalar-only variant achieves 'comparable pointwise reconstruction error'; providing the specific MSE values for both models would allow readers to assess the magnitude of the difference.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their positive summary of the technical contributions and for highlighting the important issue of evaluation scope. We address the major comment below with a commitment to textual revisions that better qualify our claims.

read point-by-point responses

  1. Referee: The reported results and the claim of an 'accurate and efficient surrogate for XANES simulation' are based entirely on performance within the narrow domain of FDMNES-simulated iron oxide surface facets. No evaluation on experimental XANES spectra or on materials outside iron oxides is provided. This untested sim-to-real and cross-chemistry transfer is load-bearing for the broader utility claimed in the abstract's final sentence, as failure to generalize would limit the model's applicability to ML-assisted spectroscopy.

    Authors: We agree that the evaluation is restricted to FDMNES simulations of iron oxide surface facets and provides no experimental spectra or cross-chemistry results. This is a substantive limitation for the broader surrogate and ML-assisted spectroscopy claims. The manuscript frames the work as a controlled demonstration on a high-fidelity simulation dataset chosen for its relevance to surface chemistry, enabling rigorous ablations of the equivariant components and derivative loss. To address the concern, we will revise the abstract to explicitly limit the reported accuracy and surrogate utility to the iron-oxide simulation setting, and we will add a limitations section that acknowledges the absence of sim-to-real and cross-material validation while outlining planned future extensions. These changes will align the stated scope with the evidence presented without altering the technical results or ablations. revision: partial

Circularity Check

0 steps flagged

No significant circularity; evaluation uses independent held-out simulations

full rationale

The paper trains and evaluates an E(3)-equivariant GNN on a fixed dataset of 5,941 FDMNES simulations of iron-oxide facets, reporting test-set MSE of 1.0e-3 together with qualitative reproduction of edge, pre-edge and oscillation features. No equation in the architecture (tensor-product message passing, spherical-harmonic edges, absorber-query attention, custom normalization, gated residuals, multi-scale Gaussian readout) is shown to be algebraically identical to its inputs or to a fitted parameter that is then relabeled as a prediction. Ablation results compare variants on the same held-out split and do not reduce any reported metric to a self-defined quantity. No self-citations are invoked as load-bearing uniqueness theorems or ansatzes. The central performance claim therefore rests on external simulation data rather than on any definitional or self-referential reduction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The performance claim rests on the assumption that FDMNES simulations are faithful labels and that the listed architectural choices (attention pooling, derivative loss, background term) are responsible for the reported gains; these are not independently verified from the abstract alone.

free parameters (1)
  • neural network weights and hyperparameters
    All model parameters are fitted to the 5941-simulation training set.
axioms (1)
  • domain assumption FDMNES simulations provide accurate ground-truth XANES spectra for iron oxide facets
    All quantitative results are measured against these simulations.

pith-pipeline@v0.9.0 · 5606 in / 1310 out tokens · 53797 ms · 2026-05-10T14:55:15.880369+00:00 · methodology

discussion (0)

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

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. ChemGraph-XANES: An Agentic Framework for XANES Simulation and Analysis

    cond-mat.mtrl-sci 2026-04 unverdicted novelty 6.0

    An LLM-orchestrated framework automates the full XANES workflow from natural language to normalized spectra and curated data.

Reference graph

Works this paper leans on

42 extracted references · 2 canonical work pages · cited by 1 Pith paper

  1. [1]

    K-edge xanes investigation of fe-based oxides by density functional theory calculations.The Journal of Physical Chemistry C, 125(47):26229– 26239, 2021

    Jing Zhu, Zhenhua Zeng, and Wei-Xue Li. K-edge xanes investigation of fe-based oxides by density functional theory calculations.The Journal of Physical Chemistry C, 125(47):26229– 26239, 2021

    2021

  2. [2]

    Modelling amorphous materials via a joint solid-state nmr and x-ray absorption spectroscopy and dft approach: application to alumina.Chemical science, 14(5):1155–1167, 2023

    Angela F Harper, Steffen P Emge, Pieter CMM Magusin, Clare P Grey, and Andrew J Morris. Modelling amorphous materials via a joint solid-state nmr and x-ray absorption spectroscopy and dft approach: application to alumina.Chemical science, 14(5):1155–1167, 2023

    2023

  3. [3]

    Jiayi Xu, Colton Lund, Prajay Patel, Yu Lim Kim, and Cong Liu. Recent advances on compu- tational modeling of supported single-atom and cluster catalysts: characterization, catalyst– support interaction, and active site heterogeneity.Catalysts, 14(4):224, 2024

    2024

  4. [4]

    In situ/operando electrocatalyst characteriza- tion by x-ray absorption spectroscopy.Chemical reviews, 121(2):882–961, 2020

    Janis Timoshenko and Beatriz Roldan Cuenya. In situ/operando electrocatalyst characteriza- tion by x-ray absorption spectroscopy.Chemical reviews, 121(2):882–961, 2020

    2020

  5. [5]

    Dynamics of heterogeneous catalytic processes at operando conditions.JACS Au, 1(12):2100–2120, 2021

    Xiangcheng Shi, Xiaoyun Lin, Ran Luo, Shican Wu, Lulu Li, Zhi-Jian Zhao, and Jinlong Gong. Dynamics of heterogeneous catalytic processes at operando conditions.JACS Au, 1(12):2100–2120, 2021

    2021

  6. [6]

    Jiayi Xu, Prajay Patel, A Jeremy Kropf, David Kaphan, Massimiliano Delferro, and Cong Liu. Theoretical investigation of the hydrogenation of cyclohexene catalyzed by supported single- atom sites on redox noninnocent limn2o4 and li2mn2o4 surfaces.The Journal of Physical Chemistry C, 128(12):4946–4957, 2024

    2024

  7. [7]

    Rishu Khurana and Cong Liu. Unveiling the redox noninnocence of metallocorroles: Explor- ing k-edge x-ray absorption near-edge spectroscopy with a multiconfigurational wave function approach.The Journal of Physical Chemistry Letters, 15(44):10985–10995, 2024

    2024

  8. [8]

    Integrated experimental and computational k-edge x-ray absorption near-edge structure analysis of vanadium catalysts.The Journal of Physical Chemistry C, 126(29):11949–11962, 2022

    Prajay Patel, Zheng Lu, Mehrafshan G Jafari, Cristina Hernandez-Prieto, Pavel Zatsepin, Daniel J Mindiola, David M Kaphan, Massimiliano Delferro, A Jeremy Kropf, and Cong Liu. Integrated experimental and computational k-edge x-ray absorption near-edge structure analysis of vanadium catalysts.The Journal of Physical Chemistry C, 126(29):11949–11962, 2022

    2022

  9. [9]

    Multicode benchmark on simulated ti k-edge x-ray absorption spectra of ti-o compounds.Physical Review Materials, 8(1):013801, 2024

    Fanchen Meng, Benedikt Maurer, Fabian Peschel, Sencer Selcuk, Mark Hybertsen, Xiaohui Qu, Christian Vorwerk, Claudia Draxl, John Vinson, and Deyu Lu. Multicode benchmark on simulated ti k-edge x-ray absorption spectra of ti-o compounds.Physical Review Materials, 8(1):013801, 2024

    2024

  10. [10]

    Jiayi Xu, Prajay Patel, Di-Jia Liu, Tao Xu, and Cong Liu. Understanding the dynamic evolution of atomically dispersed cu catalyst for co2 electrochemical conversion using integrated xanes analysis and mechanistic studies.Journal of Catalysis, 425:296–305, 2023. 16

    2023

  11. [11]

    F. Neese. The orca program system.WIRES Comput. Molec. Sci., 2(1):73–78, 2012

    2012

  12. [12]

    Parameter- free calculations of x-ray spectra with feff9.Physical Chemistry Chemical Physics, 12(21):5503– 5513, 2010

    John J Rehr, Joshua J Kas, Fernando D Vila, Micah P Prange, and Kevin Jorissen. Parameter- free calculations of x-ray spectra with feff9.Physical Chemistry Chemical Physics, 12(21):5503– 5513, 2010

    2010

  13. [13]

    Self-consistent aspects of x-ray absorption calculations.Journal of Physics: Condensed Matter, 21(34):345501, 2009

    O Bun˘ au and Yves Joly. Self-consistent aspects of x-ray absorption calculations.Journal of Physics: Condensed Matter, 21(34):345501, 2009

    2009

  14. [14]

    Sergey A Guda, Alexander A Guda, Mikhail A Soldatov, Kirill A Lomachenko, Aram L Bugaev, Carlo Lamberti, Wojciech Gawelda, Christian Bressler, Grigory Smolentsev, Alexan- der V Soldatov, et al. Optimized finite difference method for the full-potential xanes simula- tions: Application to molecular adsorption geometries in mofs and metal–ligand intersystem ...

    2015

  15. [15]

    Galv´ an, Ali Alavi, Flavia Aleotti, Francesco Aquilante, Jochen Autschbach, Davide Avagliano, Alberto Baiardi, Jie J Bao, Stefano Battaglia, et al

    Giovanni Li Manni, Ignacio Fdez. Galv´ an, Ali Alavi, Flavia Aleotti, Francesco Aquilante, Jochen Autschbach, Davide Avagliano, Alberto Baiardi, Jie J Bao, Stefano Battaglia, et al. The openmolcas web: A community-driven approach to advancing computational chemistry. Journal of Chemical Theory and Computation, 19(20):6933–6991, 2023

    2023

  16. [16]

    Ab initio theory and calculations of x-ray spectra.Comptes Rendus Physique, 10(6):548–559, 2009

    John J Rehr, Joshua J Kas, Micah P Prange, Adam P Sorini, Yoshinari Takimoto, and Fer- nando Vila. Ab initio theory and calculations of x-ray spectra.Comptes Rendus Physique, 10(6):548–559, 2009

    2009

  17. [17]

    Theoretical approaches to x-ray absorption fine structure

    John J Rehr and Robert C Albers. Theoretical approaches to x-ray absorption fine structure. Reviews of modern physics, 72(3):621, 2000

    2000

  18. [18]

    Recent advances in x-ray absorption near edge structure (xanes) simulations for catalysis: Theories and applications.Annual Reports in Computational Chemistry, 20:157–187, 2024

    Jiayi Xu, Yu Lim Kim, Rishu Khurana, Shana Havenridge, Prajay Patel, and Cong Liu. Recent advances in x-ray absorption near edge structure (xanes) simulations for catalysis: Theories and applications.Annual Reports in Computational Chemistry, 20:157–187, 2024

    2024

  19. [19]

    Soft x-ray absorption spectroscopy of liquids and solutions.Chemical reviews, 117(23):13909–13934, 2017

    Jacob W Smith and Richard J Saykally. Soft x-ray absorption spectroscopy of liquids and solutions.Chemical reviews, 117(23):13909–13934, 2017

    2017

  20. [20]

    Integrating explainability into graph neural network models for the prediction of x-ray absorption spectra.Journal of the American Chemical Society, 145(41):22584–22598, 2023

    Amir Kotobi, Kanishka Singh, Daniel Hoche, Sadia Bari, Robert H Meißner, and Annika Bande. Integrating explainability into graph neural network models for the prediction of x-ray absorption spectra.Journal of the American Chemical Society, 145(41):22584–22598, 2023

    2023

  21. [21]

    Random forest machine learning models for inter- pretable x-ray absorption near-edge structure spectrum-property relationships.npj Computa- tional Materials, 6(1):109, 2020

    Steven B Torrisi, Matthew R Carbone, Brian A Rohr, Joseph H Montoya, Yang Ha, Junko Yano, Santosh K Suram, and Linda Hung. Random forest machine learning models for inter- pretable x-ray absorption near-edge structure spectrum-property relationships.npj Computa- tional Materials, 6(1):109, 2020

    2020

  22. [22]

    Understand- ing x-ray absorption spectra by means of descriptors and machine learning algorithms.npj Computational Materials, 7(1):203, 2021

    Alexander A Guda, Sergey A Guda, Andrea Martini, AN Kravtsova, Alexander Algasov, Aram Bugaev, Stanislav P Kubrin, LV Guda, Petr ˇSot, Jeroen A van Bokhoven, et al. Understand- ing x-ray absorption spectra by means of descriptors and machine learning algorithms.npj Computational Materials, 7(1):203, 2021

    2021

  23. [23]

    Interpretable multimodal machine learning analysis of x-ray absorption near-edge spectra and pair distribu- tion functions.npj Computational Materials, 11(1):98, 2025

    Tanaporn Na Narong, Zoe N Zachko, Steven B Torrisi, and Simon JL Billinge. Interpretable multimodal machine learning analysis of x-ray absorption near-edge spectra and pair distribu- tion functions.npj Computational Materials, 11(1):98, 2025. 17

    2025

  24. [24]

    Machine-learning x-ray absorption spectra to quantitative accuracy.Physical review letters, 124(15):156401, 2020

    Matthew R Carbone, Mehmet Topsakal, Deyu Lu, and Shinjae Yoo. Machine-learning x-ray absorption spectra to quantitative accuracy.Physical review letters, 124(15):156401, 2020

    2020

  25. [25]

    Accurate, affordable, and generalizable machine learning simu- lations of transition metal x-ray absorption spectra using the xanesnet deep neural network

    CD Rankine and TJ Penfold. Accurate, affordable, and generalizable machine learning simu- lations of transition metal x-ray absorption spectra using the xanesnet deep neural network. The Journal of Chemical Physics, 156(16), 2022

    2022

  26. [26]

    Deepfit: physically and chemically informed xas-structure fitting made simple.The Journal of Physical Chemistry Letters, 2026

    Kirill Kulaev, Bogdan Protsenko, Weiren Cheng, Qinghua Liu, Dmitry Gorbunov, Mikhail Li- far, Valery Vlasenko, Anatolii Burlov, Sergey Guda, Alexander Guda, et al. Deepfit: physically and chemically informed xas-structure fitting made simple.The Journal of Physical Chemistry Letters, 2026

    2026

  27. [27]

    A graph neural network-based approach to xanes data analysis.The Journal of Physical Chemistry A, 129(4):874–884, 2025

    Fei Zhan, Haodong Yao, Zhi Geng, Lirong Zheng, Can Yu, Xue Han, Xueqi Song, Shuguang Chen, and Haifeng Zhao. A graph neural network-based approach to xanes data analysis.The Journal of Physical Chemistry A, 129(4):874–884, 2025

    2025

  28. [28]

    Nugraph2 with context-aware inputs: physics-inspired improvements in semantic segmentation.Journal of High Energy Physics, 2026(3):149, 2026

    Vitor F Grizzi, Margaret Voetberg, V Hewes, Giuseppe Cerati, and Hadi Meidani. Nugraph2 with context-aware inputs: physics-inspired improvements in semantic segmentation.Journal of High Energy Physics, 2026(3):149, 2026

    2026

  29. [29]

    Robust machine learning inference from x-ray absorption near edge spectra through featurization.Chemistry of Materials, 36(5):2304–2313, 2024

    Yiming Chen, Chi Chen, Inhui Hwang, Michael J Davis, Wanli Yang, Chengjun Sun, Gi- Hyeok Lee, Dylan McReynolds, Daniel Allan, Juan Marulanda Arias, et al. Robust machine learning inference from x-ray absorption near edge spectra through featurization.Chemistry of Materials, 36(5):2304–2313, 2024

    2024

  30. [30]

    Graph neural networks for materials science and chemistry.Communications Materials, 3(1):93, 2022

    Patrick Reiser, Marlen Neubert, Andr´ e Eberhard, Luca Torresi, Chen Zhou, Chen Shao, Hous- sam Metni, Clint van Hoesel, Henrik Schopmans, Timo Sommer, et al. Graph neural networks for materials science and chemistry.Communications Materials, 3(1):93, 2022

    2022

  31. [31]

    Enhanced exploration of lif–naf thermal con- ductivity through transferable equivariant graph neural networks.Journal of Applied Physics, 137(6), 2025

    Luca Murg, Shao-Chun Lee, Vitor F Grizzi, et al. Enhanced exploration of lif–naf thermal con- ductivity through transferable equivariant graph neural networks.Journal of Applied Physics, 137(6), 2025

    2025

  32. [32]

    and Joshi, Chaitanya K and Schmidt, Victor and Miret, Santiago and Malliaros, Fragkiskos D

    Alexandre Duval, Simon V Mathis, Chaitanya K Joshi, Victor Schmidt, Santiago Miret, Fragkiskos D Malliaros, Taco Cohen, Pietro Lio, Yoshua Bengio, and Michael Bronstein. A hitchhiker’s guide to geometric gnns for 3d atomic systems.arXiv preprint arXiv:2312.07511, 2023

    work page arXiv 2023

  33. [33]

    Geiger and T

    Mario Geiger and Tess Smidt. e3nn: Euclidean neural networks.arXiv preprint arXiv:2207.09453, 2022

    work page arXiv 2022

  34. [34]

    Fe k-edge x-ray absorption spectroscopy study of nanosized nominal magnetite.The Journal of Physical Chemistry C, 118(2):1332–1346, 2014

    Cristina Piquer, MA Laguna-Marco, Alejandro G Roca, R Boada, C Guglieri, and Jes´ us Chaboy. Fe k-edge x-ray absorption spectroscopy study of nanosized nominal magnetite.The Journal of Physical Chemistry C, 118(2):1332–1346, 2014

    2014

  35. [35]

    Electrosynthesis of urea by using fe2o3 nanoparticles encapsulated in a conductive metal–organic framework.Nature Synthesis, 3(11):1404–1413, 2024

    Da-Shuai Huang, Xiao-Feng Qiu, Jia-Run Huang, Min Mao, Lingmei Liu, Yu Han, Zhen-Hua Zhao, Pei-Qin Liao, and Xiao-Ming Chen. Electrosynthesis of urea by using fe2o3 nanoparticles encapsulated in a conductive metal–organic framework.Nature Synthesis, 3(11):1404–1413, 2024

    2024

  36. [36]

    18 Persson

    Anubhav Jain, Shyue Ping Ong, Geoffroy Hautier, Wei Chen, William Davidson Richards, Stephen Dacek, Shreyas Cholia, Dan Gunter, David Skinner, Gerbrand Ceder, and Kristin A. 18 Persson. Commentary: The materials project: A materials genome approach to accelerating materials innovation.APL Materials, 1(1):011002, 07 2013

    2013

  37. [37]

    Persson, and Shyue Ping Ong

    Richard Tran, Zihan Xu, Balachandran Radhakrishnan, Donald Winston, Wenhao Sun, Kristin A. Persson, and Shyue Ping Ong. Surface energies of elemental crystals.Scientific Data, 3(1):160080, Sep 2016

    2016

  38. [38]

    Wood, Misko Dzamba, Xiang Fu, Meng Gao, Muhammed Shuaibi, Luis Barroso- Luque, Kareem Abdelmaqsoud, Vahe Gharakhanyan, John R

    Brandon M. Wood, Misko Dzamba, Xiang Fu, Meng Gao, Muhammed Shuaibi, Luis Barroso- Luque, Kareem Abdelmaqsoud, Vahe Gharakhanyan, John R. Kitchin, Daniel S. Levine, Kyle Michel, Anuroop Sriram, Taco Cohen, Abhishek Das, Ammar Rizvi, Sushree Jagriti Sahoo, Zachary W. Ulissi, and C. Lawrence Zitnick. Uma: A family of universal models for atoms, 2025

    2025

  39. [39]

    Wood, Misko Dzamba, Meng Gao, Ammar Rizvi, C

    Luis Barroso-Luque, Muhammed Shuaibi, Xiang Fu, Brandon M. Wood, Misko Dzamba, Meng Gao, Ammar Rizvi, C. Lawrence Zitnick, and Zachary W. Ulissi. Open materials 2024 (omat24) inorganic materials dataset and models, 2024

    2024

  40. [40]

    Y. Joly. X-ray absorption near-edge structure calculations beyond the muffin-tin approxima- tion.Phys. Rev. B, 63:125120, Mar 2001

    2001

  41. [41]

    Curriculum learning

    Yoshua Bengio, J´ erˆ ome Louradour, Ronan Collobert, and Jason Weston. Curriculum learning. InProceedings of the 26th annual international conference on machine learning, pages 41–48, 2009

    2009

  42. [42]

    Multi-task learning using uncertainty to weigh losses for scene geometry and semantics

    Alex Kendall, Yarin Gal, and Roberto Cipolla. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 7482–7491, 2018. 19

    2018