Pith. sign in

REVIEW 4 major objections 5 minor 55 references

EquiHGNN: Scalable Rotationally Equivariant Hypergraph Neural Networks

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read EquiHGNN shows hypergraph structure plus rotation-aware geometric embeddings outperforms 2D graph models on large molecules.

desk verdict Useful empirical study on equivariant backbones plus hypergraph pooling, but the abstract's design rule is contradicted by its own Molecule3D table and the equivariance claim is asserted, not proven. read the letter →

arxiv 2505.05650 v1 pith:DP2ZAA3V submitted 2025-05-08 cs.LG

classification cs.LG
keywords hypergraphneuralnetworksrotationequivariancemolecularpropertypredictiongeometricdeeplearninghigher-orderinteractionsconjugatedbondslarge-scaledatasetsQM9
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 argues that the best way to combine geometric and high-order structure for molecular property prediction is simpler than it sounds: embed atoms with an equivariant geometric model, then feed those embeddings into a standard hypergraph neural network whose hyperedges are conjugated bonds. The experiments claim that hypergraph message passing alone gives little on small molecules such as QM9 and OPV monomers, but consistently beats 2D GNNs on large molecules such as PCQM4Mv2, Molecule3D, and OPV polymers. Adding 3D geometric features on top of hypergraphs improves accuracy further on most datasets. If the pattern holds, practitioners get a modular, plug-and-play recipe: reuse a strong equivariant backbone, add a hypergraph readout, and scale to large chemical datasets without designing custom equivariant topological layers.

What carries the argument

The core mechanism is a two-stage pipeline: a rotation-equivariant backbone first produces symmetry-aware atom embeddings from 3D coordinates, and those embeddings initialize the node features of an AllSet-style hypergraph neural network, which models conjugated bonds as hyperedges and updates representations through two permutation-invariant set functions, one from nodes to hyperedges and one back. This carries the argument by turning the hard problem of equivariant hypergraph message passing into a feature-initialization problem, so the topology module stays standard and scalable while the geometry enters through the input representation.

What would settle it

Run the trained EquiHGNN on rotated copies of the same molecules and check whether the prediction changes by the corresponding rotation or stays identical; if it stays identical, the model is invariant and the equivariance claim is unsupported. A second check is to rerun PCQM4Mv2 with 3D coordinates available for the entire test set, since the current setup only has coordinates for the training subset.

Watch

Extended reading notes

Core claim

EquiHGNN's central claim is that you can get most of the benefit of equivariant geometric deep learning in a hypergraph model without designing an equivariant hypergraph message-passing layer: take an equivariant backbone to produce symmetry-aware atom embeddings, combine them with invariant scalar features, and use those as initial node features in the AllSet/MHNN hypergraph framework with conjugated bonds as hyperedges. The paper reports that this hybrid consistently beats 2D GNNs and the non-geometric MHNN baseline on large molecules, and that adding geometry further improves accuracy on QM9, OPV, and PCQM4Mv2. The one reported exception is Molecule3D, where plain MHNN edges out the geometric EGNN-MHNN variant. The authors read the overall pattern as evidence that high-order interactions matter most at scale and that spatial information is a complementary signal.

Load-bearing premise

The whole pipeline is only as equivariant as the backbone that produces the initial atom embeddings; the hypergraph message-passing stage is a standard invariant layer with no defined rotation action, so if the backbone emits invariant scalars, the model is invariant rather than rotationally equivariant.

Editorial extensions

If this is right

  • On large molecules, hypergraph models beat 2D GNNs, so high-order interactions become more valuable as molecular size grows.
  • Adding geometric features on top of hypergraphs further lowers MAE on QM9, OPV, and PCQM4Mv2, so spatial information and hyperedges complement each other.
  • Equivariant embeddings can be injected at initialization rather than by redesigning hypergraph message passing, yielding a modular recipe that reuses existing public backbones.
  • On small molecules such as QM9, plain hypergraph models underperform 2D GNNs; geometric features are needed to realize gains.
  • The Molecule3D result shows the geometric benefit is not universal, so the design rule likely depends on dataset characteristics such as conformational flexibility or coordinate reliability.

Reading between the lines

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

  • Because the equivariant information enters only through initial node features, the same recipe could transfer to other invariant topological networks: precompute embeddings from a strong equivariant backbone, freeze or fine-tune them, and keep the topological layer unchanged.
  • The Molecule3D result (plain MHNN 117.55 meV versus EGNN-MHNN 122.25 meV) suggests a boundary condition worth testing: geometric features may hurt when coordinates are noisy or when the dataset's property depends mostly on topology.
  • A direct rotation test on the full pipeline would settle whether 'rotationally equivariant' applies to the whole model or only to the backbone; if the backbone outputs invariant scalar embeddings, the hybrid is invariant, not equivariant.
  • The hyperedge construction is limited to conjugated bonds, and the paper notes rings performed worse; a systematic comparison of chemically meaningful high-order units could map when hypergraph structure helps most.
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

4 major / 5 minor

Summary. The paper proposes EquiHGNN, a framework that combines an equivariant geometric backbone (EGNN, FAFormer, or Equiformer) with a hypergraph neural network (MHNN/AllSet) for molecular property prediction. The authors construct hyperedges from conjugated bonds and initialize hypergraph node features with backbone embeddings plus scalar features. They report experiments on QM9, OPV, PCQM4Mv2, and Molecule3D and claim that high-order interactions give limited benefits on small molecules but consistently outperform 2D graphs on larger molecules, and that adding geometric features to high-order structures further improves performance. The paper also makes a methodological claim that the resulting pipeline is rotationally equivariant.

Significance. If the empirical pattern were robust, the paper would provide a useful design rule: hypergraph structure matters most for large molecules and 3D geometry adds value on top of high-order representations. The plug-and-play architecture is simple and could be a practical baseline for combining equivariant backbones with hypergraph message passing. The manuscript also has strengths: it evaluates on multiple public benchmarks, reports error bars, and ships code at a public repository. However, the central empirical claim is contradicted by the paper's own Table 5 and partly by Table 3, and the equivariance claim is asserted without a group-action definition or proof. These issues are load-bearing for the title and abstract, so the current version overstates what is demonstrated.

major comments (4)
  1. [§4.3.4, Table 5] The abstract and Section 1 state that adding geometric features to high-order structures further improves performance, but Table 5 contradicts this on Molecule3D: MHNN achieves 117.55 ± 0.28 meV while EGNN-MHNN achieves 122.25 ± 0.24 meV, a degradation of about 4.7 meV that is well outside the reported error bars. Since the conclusion is phrased as a general design rule, this single counterexample is decisive and the claim needs to be restricted to the datasets and backbones where the improvement actually holds.
  2. [§3.2] The name "Equivariant HyperGraph Neural Network" and the statement that the design "naturally embeds equivariant information into the model" are not supported by the construction as written. The hypergraph module is the standard AllSet/MHNN with permutation-invariant set functions; no group action of SO(3) or E(3) is defined on the node or hyperedge feature spaces, and no theorem shows that the hypergraph message-passing updates commute with rotations or translations. If the backbone outputs invariant scalar embeddings, the composed model is invariant, not equivariant. The authors should either define the group action on all intermediate representations and prove equivariance preservation, or reframe the contribution as a hypergraph model with geometric feature initialization rather than a rotationally equivariant model.
  3. [Table 3, §4.3.2] The claim that high-order interactions offer limited benefits for small molecules is contradicted by the OPV molecular tasks. MHNN achieves 34.02 meV on Δε versus 50.45 and 55.8 meV for GIN and GAT, and 24.46 meV on εLUMO versus 53.29 and 46.68 meV; these are substantial improvements, not limited benefits. Conversely, on polymer tasks MHNN's advantage over 2D graphs is mixed, since GAT achieves a lower error than MHNN on polymer Δε (47.91 vs 48.95 meV). The abstract and Section 4.3.2 should be rewritten to describe the actual dataset- and task-dependent pattern rather than a uniform small-molecule/large-molecule dichotomy.
  4. [§4.1, §4.3.3, §4.3.4] The large-scale comparisons rest on nonstandard evaluation protocols. PCQM4Mv2 experiments use only the training subset where 3D structures are available, with a custom 80-10-10 split, rather than the official PCQM4Mv2 train/validation/test split; Molecule3D is likewise split 80-10-10 with only one equivariant variant (EGNN-MHNN) evaluated on both large datasets. Consequently, the claim that high-order interactions "consistently outperform 2D graphs on larger ones" is based on custom splits and a single backbone, and the results are not directly comparable to standard benchmark numbers. Please specify the exact comparison protocol and test whether the conclusion holds under standard splits and with more than one equivariant backbone.
minor comments (5)
  1. [§2.3, §3.1, §4.3.4] There are typos that should be corrected: "symetry" in Section 2.3, "prediciotn" and "hidene" in Section 3.1, and "PCMQM4Mv2" in the Table 4 caption.
  2. [§4.3] In the paragraph before Table 2, the citation for GAT appears as "[?]" and should be resolved to reference [37].
  3. [Table 2] The Equiformer-MHNN row is misformatted for the Δε and ⟨R²⟩ columns; the values "230.77±1.8 102815±445.433" need to be split clearly so the table can be parsed.
  4. [§4.3.4] The phrase "slightly higher error" understates the EGNN-MHNN result on Molecule3D; the difference is about 4.7 meV and is outside the overlapping error bars, so the text should acknowledge that geometric features degraded accuracy on this benchmark.
  5. [§5, §6] The limitations section reports that rings performed "significantly worse" than conjugated bonds but gives no numbers; adding this result would make the hyperedge-construction discussion more informative.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: EquiHGNN is an empirical composition of external equivariant backbones with AllSet, and its central claims rest on held-out benchmarks rather than on self-referential derivations.

full rationale

EquiHGNN's forward model is an explicit pipeline: an equivariant backbone (EGNN, Equiformer, or FAFormer) produces node embeddings that are fed into the standard AllSet/MHNN hypergraph module, followed by MLP readout. No equation in the paper defines the target output in terms of the fitted labels themselves, and no fitted parameter is renamed as a prediction; the reported numbers are held-out test MAEs on public benchmarks (QM9, OPV, PCQM4Mv2, Molecule3D). The self-citations (refs 19, 40, 48 involve author Truong-Son Hy) appear only in related-work and background contexts, and none provides a uniqueness theorem, a forced ansatz, or a load-bearing premise for the empirical conclusions. The strongest-claim inconsistency flagged by the skeptic, namely that Table 5 on Molecule3D shows MHNN at 117.55 meV while EGNN-MHNN is worse at 122.25 meV, is a consistency/evidence problem, not a circular-derivation problem. Likewise, the assertion in Section 3.2 that initializing hypergraph features with equivariant embeddings preserves equivariance is underproved, but an unsupported claim is not a circular reduction. Thus the paper has no significant circularity; the score of 2 reflects only the presence of minor, non-load-bearing self-citations.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. Its central claim rests on a few domain assumptions (conjugated bonds as hyperedges, DFT labels as truth, equivariant embeddings being useful) and on one unproven structural assumption that the full hypergraph pipeline remains equivariant. The free parameters are standard ML hyperparameters plus a non-standard data split choice.

free parameters (5)
  • cutoff radius = 5 Angstrom
    Chosen as a balance between expressiveness and computational cost for equivariant backbones (Section 4.2); all geometric models use this value and results depend on it.
  • number of neighbors = 16
    Fixed neighborhood size for EGNN, FAFormer, and Equiformer (Section 4.2); affects which atoms interact within the cutoff.
  • hidden dimension = 256
    Hidden size for all models in the architecture overview (Table 6); not tuned per dataset.
  • learning rate and epochs = 1e-4, 400
    Fixed training schedule for all experiments (Section 4.2).
  • data split ratio = 80-10-10
    Random split for all datasets; for PCQM4Mv2 and Molecule3D this is a split of the training subset, not the official test split (Section 4.1).
assumptions (4)
  • domain assumption RDKit-detected conjugated bonds are appropriate hyperedges for molecular property prediction
    Section 4.1 states hyperedges are conjugated bonds; the whole empirical study depends on this choice.
  • domain assumption Equivariant node embeddings from EGNN/FAFormer/Equiformer retain sufficient information for the downstream hypergraph model
    Section 3.2 assumes that initializing hypergraph features with these embeddings yields better representations; no ablation separates the equivariance benefit from added parameters.
  • domain assumption DFT-computed properties are the ground truth labels
    Section 4 relies on QM9, OPV, PCQM4Mv2, and Molecule3D labels as computed by DFT.
  • ad hoc to paper The hypergraph message passing of AllSet/MHNN preserves the geometric equivariance of its input features
    This is the paper's implicit assumption behind the title claim; it is invoked in Section 3.2 but never proven.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EquiHGNN: Scalable Rotationally Equivariant Hypergraph Neural Networks." pith.science (2026). https://pith.science/paper/DP2ZAA3V

@misc{pith2026250505650,
  author       = {Pith},
  title        = {Pith review of: EquiHGNN: Scalable Rotationally Equivariant Hypergraph Neural Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DP2ZAA3V}},
  note         = {Machine review of arXiv:2505.05650}
}
read the original abstract

Molecular interactions often involve high-order relationships that cannot be fully captured by traditional graph-based models limited to pairwise connections. Hypergraphs naturally extend graphs by enabling multi-way interactions, making them well-suited for modeling complex molecular systems. In this work, we introduce EquiHGNN, an Equivariant HyperGraph Neural Network framework that integrates symmetry-aware representations to improve molecular modeling. By enforcing the equivariance under relevant transformation groups, our approach preserves geometric and topological properties, leading to more robust and physically meaningful representations. We examine a range of equivariant architectures and demonstrate that integrating symmetry constraints leads to notable performance gains on large-scale molecular datasets. Experiments on both small and large molecules show that high-order interactions offer limited benefits for small molecules but consistently outperform 2D graphs on larger ones. Adding geometric features to these high-order structures further improves the performance, emphasizing the value of spatial information in molecular learning. Our source code is available at https://github.com/HySonLab/EquiHGNN/

Figures

Figures reproduced from arXiv: 2505.05650 by the authors.

Figure 1
Figure 1. a) Illustration of a hypergraph constructed from a molecule, where vertices represent atoms and hyperedges represent conjugated bonds, highlighted in blue and orange. b) Hypergraph to Bipartite representations. interactions. This approach has already led to state-of-the-art performance in a variety of machine learning tasks [12, 15, 11, 16], and holds significant promise to advance research and applications in the a… view at source ↗
Figure 2
Figure 2. Overview of the Equivariant Hypergraph Neural Network framework. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 24 canonical work pages

  1. [1]

    Molecular representations in ai-driven drug discovery: a review and practical guide.Journal of Cheminformatics, 12(1):56, 2020

    Laurianne David, Amol Thakkar, Rocío Mercado, and Ola Engkvist. Molecular representations in ai-driven drug discovery: a review and practical guide.Journal of Cheminformatics, 12(1):56, 2020

  2. [2]

    A review of molecular representa- tion in the age of machine learning.Wiley Interdisciplinary Reviews: Computational Molecular Science, 12(5):e1603, 2022

    Daniel S Wigh, Jonathan M Goodman, and Alexei A Lapkin. A review of molecular representa- tion in the age of machine learning.Wiley Interdisciplinary Reviews: Computational Molecular Science, 12(5):e1603, 2022

  3. [3]

    Neural message passing for quantum chemistry

    Justin Gilmer, Samuel S Schoenholz, Patrick F Riley, Oriol Vinyals, and George E Dahl. Neural message passing for quantum chemistry. InInternational conference on machine learning, pages 1263–1272. PMLR, 2017

  4. [4]

    Modeling polypharmacy side effects with graph convolutional networks.Bioinformatics, 34(13):i457–i466, 2018

    Marinka Zitnik, Monica Agrawal, and Jure Leskovec. Modeling polypharmacy side effects with graph convolutional networks.Bioinformatics, 34(13):i457–i466, 2018

  5. [5]

    Structure-based protein function prediction using graph convolutional networks.Nature communications, 12(1):3168, 2021

    Vladimir Gligorijevi´c, P Douglas Renfrew, Tomasz Kosciolek, Julia Koehler Leman, Daniel Berenberg, Tommi Vatanen, Chris Chandler, Bryn C Taylor, Ian M Fisk, Hera Vlamakis, et al. Structure-based protein function prediction using graph convolutional networks.Nature communications, 12(1):3168, 2021

  6. [7]

    Bronstein, Gunnar E

    Theodore Papamarkou, Tolga Birdal, Michael M. Bronstein, Gunnar E. Carlsson, Justin Curry, Yue Gao, Mustafa Hajij, Roland Kwitt, Pietro Lio, Paolo Di Lorenzo, Vasileios Maroulas, Nina Miolane, Farzana Nasrin, Karthikeyan Natesan Ramamurthy, Bastian Rieck, Simone Scardapane, Michael T Schaub, Petar Veliˇckovi´c, Bei Wang, Yusu Wang, Guowei Wei, and Ghada Z...

  7. [8]

    Architectures of topolog- ical deep learning: A survey of message-passing topological neural networks.arXiv preprint arXiv:2304.10031, 2023

    Mathilde Papillon, Sophia Sanborn, Mustafa Hajij, and Nina Miolane. Architectures of topolog- ical deep learning: A survey of message-passing topological neural networks.arXiv preprint arXiv:2304.10031, 2023

  8. [9]

    Weisfeiler and lehman go topological: Message passing simplicial networks

    Cristian Bodnar, Fabrizio Frasca, Yuguang Wang, Nina Otter, Guido F Montufar, Pietro Lio, and Michael Bronstein. Weisfeiler and lehman go topological: Message passing simplicial networks. InInternational Conference on Machine Learning, pages 1026–1037. PMLR, 2021

Show all 55 references
  1. [10]

    Weisfeiler and lehman go cellular: Cw networks.Advances in neural information processing systems, 34:2625–2640, 2021

    Cristian Bodnar, Fabrizio Frasca, Nina Otter, Yuguang Wang, Pietro Lio, Guido F Montufar, and Michael Bronstein. Weisfeiler and lehman go cellular: Cw networks.Advances in neural information processing systems, 34:2625–2640, 2021

  2. [11]

    Higher-order attention networks.arXiv preprint arXiv:2206.00606, 2(3):4, 2022

    Mustafa Hajij, Ghada Zamzmi, Theodore Papamarkou, Nina Miolane, Aldo Guzmán-Sáenz, and Karthikeyan Natesan Ramamurthy. Higher-order attention networks.arXiv preprint arXiv:2206.00606, 2(3):4, 2022

  3. [12]

    Hnhn: Hypergraph networks with hyperedge neurons.ICML Graph Representation Learning and Beyond Workshop, 2020

    Yihe Dong, Will Sawin, and Yoshua Bengio. Hnhn: Hypergraph networks with hyperedge neurons.ICML Graph Representation Learning and Beyond Workshop, 2020

  4. [13]

    You are allset: A multiset function framework for hypergraph neural networks.arXiv preprint arXiv:2106.13264, 2021

    Eli Chien, Chao Pan, Jianhao Peng, and Olgica Milenkovic. You are allset: A multiset function framework for hypergraph neural networks.arXiv preprint arXiv:2106.13264, 2021

  5. [14]

    Molecular hypergraph neural networks.The Journal of Chemical Physics, 160(14):144307, 2024

    Junwu Chen and Philippe Schwaller. Molecular hypergraph neural networks.The Journal of Chemical Physics, 160(14):144307, 2024

  6. [15]

    Topological signal processing over simplicial complexes.IEEE Transactions on Signal Processing, 68:2992–3007, 2020

    Sergio Barbarossa and Stefania Sardellitti. Topological signal processing over simplicial complexes.IEEE Transactions on Signal Processing, 68:2992–3007, 2020. 10

  7. [16]

    Bscnets: Block simplicial complex neural networks

    Yuzhou Chen, Yulia R Gel, and H Vincent Poor. Bscnets: Block simplicial complex neural networks. InProceedings of the aaai conference on artificial intelligence, volume 36, pages 6333–6341, 2022

  8. [17]

    Geometric deep learning: Grids, groups, graphs, geodesics, and gauges.arXiv preprint arXiv:2104.13478, 2021

    Michael M Bronstein, Joan Bruna, Taco Cohen, and Petar Veliˇckovi´c. Geometric deep learning: Grids, groups, graphs, geodesics, and gauges.arXiv preprint arXiv:2104.13478, 2021

  9. [18]

    Schnet: A continuous-filter convolutional neural network for modeling quantum interactions.Advances in neural information processing systems, 30, 2017

    Kristof Schütt, Pieter-Jan Kindermans, Huziel Enoc Sauceda Felix, Stefan Chmiela, Alexandre Tkatchenko, and Klaus-Robert Müller. Schnet: A continuous-filter convolutional neural network for modeling quantum interactions.Advances in neural information processing systems, 30, 2017

  10. [19]

    Cormorant: Covariant molecular neural networks

    Brandon Anderson, Truong Son Hy, and Risi Kondor. Cormorant: Covariant molecular neural networks. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox, and R. Garnett, editors,Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc., 2019

  11. [20]

    E (3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials.Nature communications, 13(1):2453, 2022

    Simon Batzner, Albert Musaelian, Lixin Sun, Mario Geiger, Jonathan P Mailoa, Mordechai Kornbluth, Nicola Molinari, Tess E Smidt, and Boris Kozinsky. E (3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials.Nature communications, 13(1):2453, 2022

  12. [21]

    A universal framework for accurate and efficient geometric deep learning of molecular systems.Scientific Reports, 13(1):19171, 2023

    Shuo Zhang, Yang Liu, and Lei Xie. A universal framework for accurate and efficient geometric deep learning of molecular systems.Scientific Reports, 13(1):19171, 2023

  13. [22]

    Enhancing geometric representations for molecules with equivariant vector-scalar interactive message passing.Nature Communications, 15(1):313, 2024

    Yusong Wang, Tong Wang, Shaoning Li, Xinheng He, Mingyu Li, Zun Wang, Nanning Zheng, Bin Shao, and Tie-Yan Liu. Enhancing geometric representations for molecules with equivariant vector-scalar interactive message passing.Nature Communications, 15(1):313, 2024

  14. [23]

    EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations

    Yi-Lun Liao, Brandon Wood, Abhishek Das*, and Tess Smidt*. EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations. InInternational Con- ference on Learning Representations (ICLR), 2024

  15. [24]

    Protein-nucleic acid complex modeling with frame averaging transformer

    Tinglin Huang, Zhenqiao Song, Rex Ying, and Wengong Jin. Protein-nucleic acid complex modeling with frame averaging transformer. InNeural Information Processing Systems, 2024

  16. [25]

    E (n) equivariant message passing simplicial networks

    Floor Eijkelboom, Rob Hesselink, and Erik J Bekkers. E (n) equivariant message passing simplicial networks. InInternational Conference on Machine Learning, pages 9071–9081. PMLR, 2023

  17. [26]

    E (n) equivariant topological neural networks.arXiv preprint arXiv:2405.15429, 2024

    Claudio Battiloro, Mauricio Tec, George Dasoulas, Michelle Audirac, Francesca Dominici, et al. E (n) equivariant topological neural networks.arXiv preprint arXiv:2405.15429, 2024

  18. [27]

    Quantum chemistry structures and properties of 134 kilo molecules.Scientific data, 1(1):1–7, 2014

    Raghunathan Ramakrishnan, Pavlo O Dral, Matthias Rupp, and O Anatole V on Lilienfeld. Quantum chemistry structures and properties of 134 kilo molecules.Scientific data, 1(1):1–7, 2014

  19. [28]

    Message-passing neural networks for high- throughput polymer screening.The Journal of chemical physics, 150(23), 2019

    Peter C St John, Caleb Phillips, Travis W Kemper, A Nolan Wilson, Yanfei Guan, Michael F Crowley, Mark R Nimlos, and Ross E Larsen. Message-passing neural networks for high- throughput polymer screening.The Journal of chemical physics, 150(23), 2019

  20. [29]

    Ogb- lsc: A large-scale challenge for machine learning on graphs.arXiv preprint arXiv:2103.09430, 2021

    Weihua Hu, Matthias Fey, Hongyu Ren, Maho Nakata, Yuxiao Dong, and Jure Leskovec. Ogb- lsc: A large-scale challenge for machine learning on graphs.arXiv preprint arXiv:2103.09430, 2021

  21. [30]

    Molecule3d: A benchmark for predicting 3d geometries from molecular graphs.arXiv preprint arXiv:2110.01717, 2021

    Zhao Xu, Youzhi Luo, Xuan Zhang, Xinyi Xu, Yaochen Xie, Meng Liu, Kaleb Dickerson, Cheng Deng, Maho Nakata, and Shuiwang Ji. Molecule3d: A benchmark for predicting 3d geometries from molecular graphs.arXiv preprint arXiv:2110.01717, 2021

  22. [31]

    E (n) equivariant graph neural networks

    Vıctor Garcia Satorras, Emiel Hoogeboom, and Max Welling. E (n) equivariant graph neural networks. InInternational conference on machine learning, pages 9323–9332. PMLR, 2021

  23. [32]

    Equiformer: Equivariant graph attention transformer for 3d atomistic graphs

    Yi-Lun Liao and Tess Smidt. Equiformer: Equivariant graph attention transformer for 3d atomistic graphs. InInternational Conference on Learning Representations, 2023. 11

  24. [33]

    Graph neural networks: A review of methods and applications.AI open, 1:57–81, 2020

    Jie Zhou, Ganqu Cui, Shengding Hu, Zhengyan Zhang, Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, and Maosong Sun. Graph neural networks: A review of methods and applications.AI open, 1:57–81, 2020

  25. [34]

    Inductive representation learning on large graphs.Advances in neural information processing systems, 30, 2017

    Will Hamilton, Zhitao Ying, and Jure Leskovec. Inductive representation learning on large graphs.Advances in neural information processing systems, 30, 2017

  26. [35]

    Semi-supervised classification with graph convolutional networks.arXiv preprint arXiv:1609.02907, 2016

    Thomas N Kipf and Max Welling. Semi-supervised classification with graph convolutional networks.arXiv preprint arXiv:1609.02907, 2016

  27. [36]

    How powerful are graph neural networks?arXiv preprint arXiv:1810.00826, 2018

    Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks?arXiv preprint arXiv:1810.00826, 2018

  28. [37]

    Graph attention networks

    Petar Veliˇckovi´c, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. Graph attention networks. InInternational Conference on Learning Representations, 2018

  29. [38]

    Directional message passing for molecular graphs

    Johannes Gasteiger, Janek Groß, and Stephan Günnemann. Directional message passing for molecular graphs. InInternational Conference on Learning Representations (ICLR), 2020

  30. [39]

    Gemnet: Universal directional graph neural networks for molecules.Advances in Neural Information Processing Systems, 34:6790–6802, 2021

    Johannes Gasteiger, Florian Becker, and Stephan Günnemann. Gemnet: Universal directional graph neural networks for molecules.Advances in Neural Information Processing Systems, 34:6790–6802, 2021

  31. [40]

    Graph representation learning, deep generative models on graphs, group equivariant molecular neural networks and multiresolution machine learning

    Truong Son Hy. Graph representation learning, deep generative models on graphs, group equivariant molecular neural networks and multiresolution machine learning. page 366

  32. [41]

    Frame averaging for invariant and equivariant network design.arXiv preprint arXiv:2110.03336, 2021

    Omri Puny, Matan Atzmon, Heli Ben-Hamu, Ishan Misra, Aditya Grover, Edward J Smith, and Yaron Lipman. Frame averaging for invariant and equivariant network design.arXiv preprint arXiv:2110.03336, 2021

  33. [42]

    Faenet: Frame averaging equivariant gnn for materials modeling

    Alexandre Agm Duval, Victor Schmidt, Alex Hernández-Garcıa, Santiago Miret, Fragkiskos D Malliaros, Yoshua Bengio, and David Rolnick. Faenet: Frame averaging equivariant gnn for materials modeling. InInternational Conference on Machine Learning, pages 9013–9033. PMLR, 2023

  34. [43]

    Cohen, Mario Geiger, Jonas Köhler, and Max Welling

    Taco S. Cohen, Mario Geiger, Jonas Köhler, and Max Welling. Spherical CNNs. InInternational Conference on Learning Representations, 2018

  35. [44]

    Tensor field networks: Rotation-and translation-equivariant neural networks for 3d point clouds.arXiv preprint arXiv:1802.08219, 2018

    Nathaniel Thomas, Tess Smidt, Steven Kearnes, Lusann Yang, Li Li, Kai Kohlhoff, and Patrick Riley. Tensor field networks: Rotation-and translation-equivariant neural networks for 3d point clouds.arXiv preprint arXiv:1802.08219, 2018

  36. [45]

    Se (3)-transformers: 3d roto-translation equivariant attention networks.Advances in neural information processing systems, 33:1970–1981, 2020

    Fabian Fuchs, Daniel Worrall, V olker Fischer, and Max Welling. Se (3)-transformers: 3d roto-translation equivariant attention networks.Advances in neural information processing systems, 33:1970–1981, 2020

  37. [46]

    Geometric and physical quantities improve e(3) equivariant message passing

    Johannes Brandstetter, Rob Hesselink, Elise van der Pol, Erik J Bekkers, and Max Welling. Geometric and physical quantities improve e(3) equivariant message passing. InInternational Conference on Learning Representations, 2022

  38. [47]

    On the expressive power of geometric graph neural networks

    Chaitanya K Joshi, Cristian Bodnar, Simon V Mathis, Taco Cohen, and Pietro Lio. On the expressive power of geometric graph neural networks. InInternational conference on machine learning, pages 15330–15355. PMLR, 2023

  39. [48]

    Anderson, and Risi Kondor

    Truong Son Hy, Shubhendu Trivedi, Horace Pan, Brandon M. Anderson, and Risi Kondor. Predicting molecular properties with covariant compositional networks.The Journal of Chemical Physics, 148(24):241745, 06 2018

  40. [49]

    Hypergraph neural networks

    Yifan Feng, Haoxuan You, Zizhao Zhang, Rongrong Ji, and Yue Gao. Hypergraph neural networks. InProceedings of the AAAI conference on artificial intelligence, volume 33, pages 3558–3565, 2019

  41. [50]

    Clifford group equivariant simplicial message passing networks.arXiv preprint arXiv:2402.10011, 2024

    Cong Liu, David Ruhe, Floor Eijkelboom, and Patrick Forré. Clifford group equivariant simplicial message passing networks.arXiv preprint arXiv:2402.10011, 2024. 12

  42. [51]

    Pubchemqc project: a large-scale first-principles electronic structure database for data-driven chemistry.Journal of chemical information and modeling, 57(6):1300–1308, 2017

    Maho Nakata and Tomomi Shimazaki. Pubchemqc project: a large-scale first-principles electronic structure database for data-driven chemistry.Journal of chemical information and modeling, 57(6):1300–1308, 2017

  43. [52]

    Scaling graph neural networks to large proteins.Journal of Chemical Theory and Computation, 2024

    Justin Airas and Bin Zhang. Scaling graph neural networks to large proteins.Journal of Chemical Theory and Computation, 2024

  44. [53]

    Matthias Fey and Jan E. Lenssen. Fast graph representation learning with PyTorch Geometric. InICLR Workshop on Representation Learning on Graphs and Manifolds, 2019

  45. [54]

    Se3set: Harnessing equivariant hypergraph neural networks for molecular representation learning.arXiv preprint arXiv:2405.16511, 2024

    Hongfei Wu, Lijun Wu, Guoqing Liu, Zhirong Liu, Bin Shao, and Zun Wang. Se3set: Harnessing equivariant hypergraph neural networks for molecular representation learning.arXiv preprint arXiv:2405.16511, 2024

  46. [55]

    Hypergraph geometry reflects higher-order dynamics in protein interaction networks.Scientific Reports, 12(1):20879, 2022

    Kevin A Murgas, Emil Saucan, and Romeil Sandhu. Hypergraph geometry reflects higher-order dynamics in protein interaction networks.Scientific Reports, 12(1):20879, 2022

  47. [56]

    Predicting ret- rosynthetic pathways using transformer-based models and a hyper-graph exploration strategy

    Philippe Schwaller, Riccardo Petraglia, Valerio Zullo, Vishnu H Nair, Rico Andreas Haeusel- mann, Riccardo Pisoni, Costas Bekas, Anna Iuliano, and Teodoro Laino. Predicting ret- rosynthetic pathways using transformer-based models and a hyper-graph exploration strategy. Chemica...

Pith tools

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