mlsanityData to 14 Oct 2026

Chi-Geometry: A Library for Benchmarking Chirality Prediction of GNNs

arXiv 2508.09097 ↗First posted 12 Aug 2025cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We introduce Chi-Geometry - a library that generates graph data for testing and benchmarking GNNs' ability to predict chirality. Chi-Geometry generates synthetic graph samples with (i) user-specified geometric and topological traits to isolate certain types of samples and (ii) randomized node positions and species to minimize extraneous correlations. Each generated graph contains exactly one chiral center labeled either R or S, while all other nodes are labeled N/A (non-chiral). The generated samples are then combined into a cohesive dataset that can be used to assess a GNN's ability to predict chirality as a node classification task. Chi-Geometry allows more interpretable and less confounding benchmarking of GNNs for prediction of chirality in the graph samples which can guide the design of new GNN architectures with improved predictive performance. We illustrate Chi-Geometry's efficacy by using it to generate synthetic datasets for benchmarking various state-of-the-art (SOTA) GNN architectures. The conclusions of these benchmarking results guided our design of two new GNN architectures. The first GNN architecture established all-to-all connections in the graph to accurately predict chirality across all challenging configurations where previously tested SOTA models failed, but at a computational cost (both for training and inference) that grows quadratically with the number of graph nodes. The second GNN architecture avoids all-to-all connections by introducing a virtual node in the original graph structure of the data, which restores the linear scaling of training and inference computational cost with respect to the number of nodes in the graph, while still ensuring competitive accuracy in detecting chirality with respect to SOTA GNN architectures.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 34 of its 49 references that are in the map. First posted on arXiv 12 Aug 2025.

Builds on

34 of 49 references.

References · 34

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2017Inductive Representation Learning on Large Graphscited by 6,320
  4. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  5. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,102
  6. 2021How Attentive are Graph Attention Networks?cited by 835
  7. 2015Convolutional Networks on Graphs for Learning Molecular Fingerprintscited by 754
  8. 2019Fast Transformer Decoding: One Write-Head is All You Needcited by 717
  9. 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 708
  10. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 694
  11. 2020On the Bottleneck of Graph Neural Networks and its Practical Implicationscited by 631
  12. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 607
22 more
  1. 2017SchNet: A continuous-filter convolutional neural network for modeling quantum interactionscited by 571
  2. 2020Directional Message Passing for Molecular Graphscited by 561
  3. 2021Understanding over-squashing and bottlenecks on graphs via curvaturecited by 478
  4. 2022MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fieldscited by 474
  5. 2018On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groupscited by 300
  6. 2022Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphscited by 264
  7. 2023A Survey on Oversmoothing in Graph Neural Networkscited by 247
  8. 2017The Reversible Residual Network: Backpropagation Without Storing Activationscited by 217
  9. 2023EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representationscited by 204
  10. 2023Zoology: Measuring and Improving Recall in Efficient Language Modelscited by 204
  11. 2020The Lipschitz Constant of Self-Attentioncited by 174
  12. 2023Quantizable Transformers: Removing Outliers by Helping Attention Heads Do Nothingcited by 151
  13. 2023Reducing SO(3) Convolutions to SO(2) for Efficient Equivariant GNNscited by 117
  14. 2024Open Materials 2024 (OMat24) Inorganic Materials Dataset and Modelscited by 52
  15. 2022Spherical Channels for Modeling Atomic Interactionscited by 47
  16. 2025On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learningcited by 43
  17. 2021Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotationscited by 29
  18. 2022Geometry-Complete Perceptron Networks for 3D Molecular Graphscited by 16
  19. 2024VN-EGNN: E(3)-Equivariant Graph Neural Networks with Virtual Nodes Enhance Protein Binding Site Identificationcited by 15
  20. 2025Towards Mechanistic Interpretability of Graph Transformers via Attention Graphscited by 13
  21. 2022Multi-task graph neural networks for simultaneous prediction of global and atomic properties in ferromagnetic systemscited by 6
  22. 2023ChiENN: Embracing Molecular Chirality with Graph Neural Networkscited by 6

Built on by

Not cited yet.