Balancing Symmetry and Efficiency in Graph Flow Matching

• URL: http://arxiv.org/abs/2602.18084v1
• Date: Fri, 20 Feb 2026 09:17:57 GMT
• Title: Balancing Symmetry and Efficiency in Graph Flow Matching
• Abstract summary: We study the trade-off for graph generative models between strict equivariance and controllable symmetry modulation.<n>We show that properly modulating the symmetry signal can delay overfitting while accelerating convergence.
• Score: 33.80974796363154
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Equivariance is central to graph generative models, as it ensures the model respects the permutation symmetry of graphs. However, strict equivariance can increase computational cost due to added architectural constraints, and can slow down convergence because the model must be consistent across a large space of possible node permutations. We study this trade-off for graph generative models. Specifically, we start from an equivariant discrete flow-matching model, and relax its equivariance during training via a controllable symmetry modulation scheme based on sinusoidal positional encodings and node permutations. Experiments first show that symmetry-breaking can accelerate early training by providing an easier learning signal, but at the expense of encouraging shortcut solutions that can cause overfitting, where the model repeatedly generates graphs that are duplicates of the training set. On the contrary, properly modulating the symmetry signal can delay overfitting while accelerating convergence, allowing the model to reach stronger performance with $19\%$ of the baseline training epochs.

Related papers

• Gromov-Monge Flow Matching for Equivariant Graph Generation [3.3603088054104284]
  In flow matching, symmetry may enter the source--target coupling.<n>We show that quotient couplings can be lifted to aligned representatives without additional cost.<n>We construct minibatch couplings using efficient Gromov--Wasserstein-type relaxations.
  arXiv  Detail & Related papers  (2026-08-27T11:01:03Z)
• Rethinking Diffusion Models with Symmetries through Canonicalization with Applications to Molecular Graph Generation [56.361076943802594]
  CanonFlow achieves state-of-the-art performance on the challenging GEOM-DRUG dataset, and the advantage remains large in few-step generation.
  arXiv  Detail & Related papers  (2026-02-16T18:58:55Z)
• Learning Time-Varying Graphs from Incomplete Graph Signals [1.7430416823420511]
  We develop an efficient Alternating Direction Multiplier algorithm for solving the problem of imputing missing data from a graph.<n>We prove that the proposed ADMM scheme converges to and we derive a stationary point.
  arXiv  Detail & Related papers  (2025-10-19T11:12:13Z)
• Implicit Augmentation from Distributional Symmetry in Turbulence Super-Resolution [2.4095472682599417]
  We show that standard convolutional neural networks (CNNs) can partially acquire rotational symmetry without explicit augmentation or specialized architectures.<n>Using 3D channel-flow with differing anisotropy, we find that models trained on more isotropic mid-plane data achieve lower equivariance error.<n>These results clarify when rotational symmetry must be explicitly incorporated into learning algorithms and when it can be obtained directly from turbulence.
  arXiv  Detail & Related papers  (2025-09-25T02:31:56Z)
• Learning (Approximately) Equivariant Networks via Constrained Optimization [25.51476313302483]
  Equivariant neural networks are designed to respect symmetries through their architecture.<n>Real-world data often departs from perfect symmetry because of noise, structural variation, measurement bias, or other symmetry-breaking effects.<n>We introduce Adaptive Constrained Equivariance (ACE), a constrained optimization approach that starts with a flexible, non-equivariant model.
  arXiv  Detail & Related papers  (2025-05-19T18:08:09Z)
• Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion [55.95767828747407]
  In domains such as molecular and protein generation, physical systems exhibit inherent symmetries that are critical to model.<n>We present a framework that reduces training variance and provides a provably lower-variance gradient estimator.<n>We also present a practical implementation of this estimator incorporating the loss and sampling procedure through a method we call Orbit Diffusion.
  arXiv  Detail & Related papers  (2025-02-14T03:26:57Z)
• IFH: a Diffusion Framework for Flexible Design of Graph Generative Models [53.219279193440734]
  Graph generative models can be classified into two prominent families: one-shot models, which generate a graph in one go, and sequential models, which generate a graph by successive additions of nodes and edges. This paper proposes a graph generative model, called Insert-Fill-Halt (IFH), that supports the specification of a sequentiality degree.
  arXiv  Detail & Related papers  (2024-08-23T16:24:40Z)
• Equivariant Denoisers Cannot Copy Graphs: Align Your Graph Diffusion Models [24.912841472542322]
  Graph diffusion models, dominant in graph generative modeling, remain underexplored for graph-to-graph translation tasks.<n>We show how alignment dramatically improves discrete diffusion model performance from 5% to a SOTA-matching 54.7% top-1 accuracy.
  arXiv  Detail & Related papers  (2024-05-27T20:57:19Z)
• Similarity Equivariant Graph Neural Networks for Homogenization of Metamaterials [3.6443770850509423]
  Soft, porous mechanical metamaterials exhibit pattern transformations that may have important applications in soft robotics, sound reduction and biomedicine.<n>We develop a machine learning-based approach that scales favorably to serve as a surrogate model.<n>We show that this network is more accurate and data-efficient than graph neural networks with fewer symmetries.
  arXiv  Detail & Related papers  (2024-04-26T12:30:32Z)
• Sparse Training of Discrete Diffusion Models for Graph Generation [45.103518022696996]
  We introduce SparseDiff, a novel diffusion model based on the observation that almost all large graphs are sparse. By selecting a subset of edges, SparseDiff effectively leverages sparse graph representations both during the noising process and within the denoising network. Our model demonstrates state-of-the-art performance across multiple metrics on both small and large datasets.
  arXiv  Detail & Related papers  (2023-11-03T16:50:26Z)
• Learning Layer-wise Equivariances Automatically using Gradients [66.81218780702125]
  Convolutions encode equivariance symmetries into neural networks leading to better generalisation performance. symmetries provide fixed hard constraints on the functions a network can represent, need to be specified in advance, and can not be adapted. Our goal is to allow flexible symmetry constraints that can automatically be learned from data using gradients.
  arXiv  Detail & Related papers  (2023-10-09T20:22:43Z)
• Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations [57.15855198512551]
  We propose a novel score-based generative model for graphs with a continuous-time framework. We show that our method is able to generate molecules that lie close to the training distribution yet do not violate the chemical valency rule.
  arXiv  Detail & Related papers  (2022-02-05T08:21:04Z)
• Permutation Invariant Graph Generation via Score-Based Generative Modeling [114.12935776726606]
  We propose a permutation invariant approach to modeling graphs, using the recent framework of score-based generative modeling. In particular, we design a permutation equivariant, multi-channel graph neural network to model the gradient of the data distribution at the input graph. For graph generation, we find that our learning approach achieves better or comparable results to existing models on benchmark datasets.
  arXiv  Detail & Related papers  (2020-03-02T03:06:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.