mlsanityData to 14 Oct 2026

NVDiff: Graph Generation through the Diffusion of Node Vectors

arXiv 2211.10794 ↗First posted 19 Nov 2022cs.LGcited by 18

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Learning to generate graphs is challenging as a graph is a set of pairwise connected, unordered nodes encoding complex combinatorial structures. Recently, several works have proposed graph generative models based on normalizing flows or score-based diffusion models. However, these models need to generate nodes and edges in parallel from the same process, whose dimensionality is unnecessarily high. We propose NVDiff, which takes the VGAE structure and uses a score-based generative model (SGM) as a flexible prior to sample node vectors. By modeling only node vectors in the latent space, NVDiff significantly reduces the dimension of the diffusion process and thus improves sampling speed. Built on the NVDiff framework, we introduce an attention-based score network capable of capturing both local and global contexts of graphs. Experiments indicate that NVDiff significantly reduces computations and can model much larger graphs than competing methods. At the same time, it achieves superior or competitive performances over various datasets compared to previous methods.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 45 of its 61 references that are in the map. First posted on arXiv 19 Nov 2022.

Builds on

45 of 61 references.

Closest ancestors · 2

Same direction, newest first

  1. 2022Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations5 Feb
  2. 2020Permutation Invariant Graph Generation via Score-Based Generative Modeling2 Mar

Other references · 43

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  4. 2013Auto-Encoding Variational Bayescited by 4,053
  5. 2015Variational Inference with Normalizing Flowscited by 2,170
  6. 2016Variational Graph Auto-Encoderscited by 1,382
  7. 2021E(n) Equivariant Graph Neural Networkscited by 965
  8. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 652
  9. 2021Score-based Generative Modeling in Latent Spacecited by 636
  10. 2020WaveGrad: Estimating Gradients for Waveform Generationcited by 622
  11. 2018GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Modelscited by 413
  12. 2021Flow Network based Generative Models for Non-Iterative Diverse Candidate Generationcited by 392
31 more
  1. 2018GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoderscited by 383
  2. 2020GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generationcited by 305
  3. 2019Efficient Graph Generation with Graph Recurrent Attention Networkscited by 234
  4. 2018Learning Deep Generative Models of Graphscited by 229
  5. 2020Hierarchical Generation of Molecular Graphs using Structural Motifscited by 208
  6. 2018Fr'echet ChemNet Distance: A metric for generative models for molecules in drug discoverycited by 194
  7. 2019Diagnosing and Enhancing VAE Modelscited by 193
  8. 2018Constrained Graph Variational Autoencoders for Molecule Designcited by 185
  9. 2020MoFlow: An Invertible Flow Model for Generating Molecular Graphscited by 185
  10. 2021GraphDF: A Discrete Flow Model for Molecular Graph Generationcited by 180
  11. 2021E(n) Equivariant Normalizing Flowscited by 165
  12. 2021Symbolic Music Generation with Diffusion Modelscited by 154
  13. 2019GraphNVP: An Invertible Flow Model for Generating Molecular Graphscited by 116
  14. 2018Graphite: Iterative Generative Modeling of Graphscited by 112
  15. 2019Graph Normalizing Flowscited by 104
  16. 2018Multi-Objective De Novo Drug Design with Conditional Graph Generative Modelcited by 101
  17. 2021D2C: Diffusion-Denoising Models for Few-shot Conditional Generationcited by 96
  18. 2021Learning Neural Generative Dynamics for Molecular Conformation Generationcited by 88
  19. 2018NeVAE: A Deep Generative Model for Molecular Graphscited by 67
  20. 2020GraphGen: A Scalable Approach to Domain-agnostic Labeled Graph Generationcited by 66
  21. 2020Scalable Deep Generative Modeling for Sparse Graphscited by 65
  22. 2018Molecular Hypergraph Grammar with its Application to Molecular Optimizationcited by 56
  23. 2019Learning Hierarchical Priors in VAEscited by 53
  24. 2022On Evaluation Metrics for Graph Generative Modelscited by 46
  25. 2020Dirichlet Graph Variational Autoencodercited by 40
  26. 2019Stochastic Blockmodels meet Graph Neural Networkscited by 39
  27. 2019A Model to Search for Synthesizable Moleculescited by 37
  28. 2020Categorical Normalizing Flows via Continuous Transformationscited by 37
  29. 2021Order Matters: Probabilistic Modeling of Node Sequence for Graph Generationcited by 33
  30. 2021Diffusion Priors In Variational Autoencoderscited by 25
  31. 2020Edge-based sequential graph generation with recurrent neural networkscited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 6; 2024: 3; 2025: 5; 2026: 4.

18 papers.

Sort:

By area:

  1. 2023Efficient and Degree-Guided Graph Generation via Discrete Diffusion Modelingcited by 80
  2. 2023A Survey on Graph Diffusion Models: Generative AI in Science for Molecule, Protein and Materialcited by 35
  3. 2025Graph Generative Pre-trained Transformercited by 17
  4. 2025Controllable Graph Generation with Diffusion Models via Inference-Time Tree Search Guidancecited by 7
  5. 2024Enhancing Diffusion-based Point Cloud Generation with Smoothness Constraintcited by 5
  6. 2023Size Matters: Large Graph Generation with HiGGscited by 3
  7. 2023EDGE++: Improved Training and Sampling of EDGEcited by 3
  8. 2023Modeling and design of heterogeneous hierarchical bioinspired spider web structures using generative deep learning and additive manufacturingcited by 2
  9. 2024Diffusion-based Graph Generative Methodscited by 2
  10. 2024Latent Conditional Diffusion-based Data Augmentation for Continuous-Time Dynamic Graph Modelcited by 2
  11. 2026Inference Attacks Against Graph Generative Diffusion Modelscited by 2
  12. 2023SANGEA: Scalable and Attributed Network Generationcited by 1
6 more
  1. 2025DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platformcited by 1
  2. 2025Incorporating Inductive Biases to Energy-based Generative Modelscited by 1
  3. 2025GraphBrep: Learning B-Rep in Graph Structure for Efficient CAD Generationcited by 0
  4. 2026Communication-efficient Federated Graph Classification via Generative Diffusion Modelingcited by 0
  5. 2026Mitigating topology biases in Graph Diffusion via Counterfactual Interventioncited by 0
  6. 2026Graph Representation via Elements of Discrete Morse and Cobordism Theoriescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Unifying Generation and Prediction on Graphs with Latent Graph Diffusion
  2. 2023Autoregressive Diffusion Model for Graph Generation
  3. 2024Neural Graph Generator: Feature-Conditioned Graph Generation using Latent Diffusion Models
  4. 2024Discrete-state Continuous-time Diffusion for Graph Generation
  5. 2022Fast Graph Generation via Spectral Diffusion
  6. 2025Generating Large Semi-Synthetic Graphs of Any Size
  7. 2023Generative Diffusion Models on Graphs: Methods and Applications
  8. 2025Do Graph Diffusion Models Accurately Capture and Generate Substructure Distributions?