Pith. sign in

REVIEW 35 cited by

DiGress: Discrete Denoising diffusion for graph generation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2209.14734 v4 pith:4Y26HST4 submitted 2022-09-29 cs.LG

Clement Vignac , Igor Krawczuk , Antoine Siraudin , Bohan Wang , Volkan Cevher , Pascal Frossard This is my paper · ORCID
classification cs.LG
keywords diffusionmodeldigressdiscreteedgegraphgraphsnode
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

This work introduces DiGress, a discrete denoising diffusion model for generating graphs with categorical node and edge attributes. Our model utilizes a discrete diffusion process that progressively edits graphs with noise, through the process of adding or removing edges and changing the categories. A graph transformer network is trained to revert this process, simplifying the problem of distribution learning over graphs into a sequence of node and edge classification tasks. We further improve sample quality by introducing a Markovian noise model that preserves the marginal distribution of node and edge types during diffusion, and by incorporating auxiliary graph-theoretic features. A procedure for conditioning the generation on graph-level features is also proposed. DiGress achieves state-of-the-art performance on molecular and non-molecular datasets, with up to 3x validity improvement on a planar graph dataset. It is also the first model to scale to the large GuacaMol dataset containing 1.3M drug-like molecules without the use of molecule-specific representations.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 35 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 71 citations worldwide. Full citation record

  1. DiffCkt: A Diffusion Model-Based Hybrid Neural Network Framework for Automatic Transistor-Level Generation of Analog Circuits

    cs.ET 2025-07 conditional novelty 7.0 of 10

    DiffCkt uses three diffusion networks to predict amplifier component counts, topology, and transistor sizes from performance specifications, and reports 2.21x to 8365x higher generation efficiency than prior analog EDA tools.

  2. Can Graph Neural Networks Learn Language with Extremely Weak Text Supervision?

    cs.LG 2024-12 conditional novelty 7.0 of 10

    Morpher adapts pre-trained GNNs to language using multi-modal prompts and a projector, achieving few-shot, cross-domain, and zero-shot unseen-class classification with weak text supervision.

  3. Graph Convolutional Attention: A Spectral Perspective on Graph Denoising and Diffusion

    cs.LG 2026-07 conditional novelty 6.5 of 10

    Under a denoising objective, linear attention is suboptimal; Graph Convolutional Attention matches idealized spectral attention on SBMs and improves graph denoising and diffusion in proportion to spectral diversity.

  4. Hierarchical Domain Generalization

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Over infinite domains, hierarchy-uniform domain generalization is impossible for every nontrivial hypothesis class; a length-generalization bound is a property of the length hierarchy, not a hierarchy-free guarantee.

  5. CoDiffGRN: Rethinking Gene Regulatory Network Inference via the BEELINE-KGC Benchmark and Co-evolutionary Discrete Diffusion

    cs.LG 2026-07 conditional novelty 6.0 of 10

    CoDiffGRN reformulates GRN inference as inductive, ranking-based knowledge graph completion and claims state-of-the-art top-K regulatory discovery on the new BEELINE-KGC benchmark.

  6. Symmetry-Structured Neural Completion of Islamic Geometric Patterns from Sparse Control Geometry

    cs.CV 2026-06 conditional novelty 6.0 of 10

    Orbit-structured graph decoding completes Islamic-style vector patterns from sparse control geometry with a constructive guarantee of exact N-fold symmetry and no measurable fidelity cost on clean synthetic data.

  7. SynCircuit: Automated Generation of New Synthetic RTL Circuits Can Enable Big Data in Circuits

    cs.LG 2025-08 conditional novelty 6.0 of 10

    SynCircuit generates new, structurally valid RTL circuits with a directed-cyclic-graph diffusion model plus post-processing and MCTS, and shows they improve ML-based PPA prediction when added to training data.

  8. Multi-domain Distribution Learning for De Novo Drug Design

    cs.LG 2025-08 conditional novelty 6.0 of 10

    DrugFlow, a flow-matching plus Markov-bridge generative model, reports state-of-the-art distributional fidelity for structure-based drug design and adds uncertainty, size adaptation, side-chain flexibility, and prefer...

  9. SBGD: Improving Graph Diffusion Generative Model via Stochastic Block Diffusion

    cs.LG 2025-08 reject novelty 6.0 of 10

    By diffusing on block subgraphs rather than the full graph, SBGD reports lower memory use and better size generalization for graph diffusion generative models.

  10. FlowMol3: Flow Matching for 3D De Novo Small-Molecule Generation

    cs.LG 2025-08 conditional novelty 6.0 of 10

    Combining self-conditioning, fake atoms, and late-stage geometry distortion lets a compact flow-matching model generate nearly always valid 3D drug-like molecules and match training-data chemistry better than existing...

  11. Beyond Interactions: Node-Level Graph Generation for Knowledge-Free Augmentation in Recommender Systems

    cs.IR 2025-07 reject novelty 6.0 of 10

    NodeDiffRec uses a two-stage diffusion process to generate pseudo-items and interactions for recommender data augmentation, reporting SOTA gains on three sparse datasets.

  12. Subgraph Generation for Generalizing on Out-of-Distribution Links

    cs.LG 2025-07 conditional novelty 6.0 of 10

    FLEX is a generative framework that synthesizes counterfactual subgraphs with a semi-implicit graph VAE and adversarially co-trains a GNN to improve out-of-distribution link prediction.

  13. DiffNMR: Diffusion Models for Nuclear Magnetic Resonance Spectra Elucidation

    physics.chem-ph 2025-07 conditional novelty 6.0 of 10

    DiffNMR uses a discrete graph diffusion model conditioned on NMR spectra to predict molecular structures, achieving 68.26% top-1 accuracy with formula on molecules up to 15 heavy atoms.

  14. GraphBrep: Learning B-Rep in Graph Structure for Efficient CAD Generation

    cs.CV 2025-07 conditional novelty 6.0 of 10

    GraphBrep replaces the redundant tree-based topology of prior B-Rep generators with an explicit graph adjacency representation, cutting training and inference cost while preserving generation quality.

  15. FreeScene: Mixed Graph Diffusion for 3D Scene Synthesis from Free Prompts

    cs.CV 2025-06 conditional novelty 6.0 of 10

    FreeScene parses free-form text and image prompts into scene graphs via a VLM-based Graph Designer, then generates 3D indoor layouts with a mixed graph diffusion transformer, reporting improved quality and controllabi...

  16. Satellites Reveal Mobility: A Commuting Origin-destination Flow Generator for Global Cities

    cs.CV 2025-05 conditional novelty 6.0 of 10

    Satellite imagery plus population is enough to generate commuting origin-destination flows that closely match models using detailed sociodemographic and point-of-interest data.

  17. A Global Commuting Origin-Destination Flow Dataset for Urban Sustainable Development

    cs.OH 2025-05 conditional novelty 6.0 of 10

    This paper releases a synthetic commuting OD flow dataset for 1,625 cities worldwide, generated by a graph diffusion model from public population, satellite, and POI data.

  18. Theoretical Benefit and Limitation of Diffusion Language Model

    cs.LG 2025-02 conditional novelty 6.0 of 10

    Masked diffusion language models have a metric-dependent efficiency tradeoff: near-optimal perplexity in constant steps, but sequence-level correctness needs linearly many steps in the worst case.

  19. Do Graph Diffusion Models Accurately Capture and Generate Substructure Distributions?

    cs.LG 2025-02 conditional novelty 6.0 of 10

    Graph diffusion models frequently distort substructure count distributions, and the paper shows theoretically and empirically that the GNN backbone's subgraph-counting expressivity controls this behavior.

  20. Categorical Schr\"odinger Bridge Matching

    cs.LG 2025-02 conditional novelty 6.0 of 10

    The paper proves that discrete-time iterative Markovian fitting converges to the Schrödinger Bridge on finite discrete spaces and introduces CSBM, a practical matching algorithm for categorical data.

  21. Graph Generative Pre-trained Transformer

    cs.LG 2025-01 conditional novelty 6.0 of 10

    G2PT represents graphs as node-then-edge token sequences and learns them with GPT-style next-token prediction, matching or beating diffusion baselines on seven graph and molecule datasets.

  22. FAP-CD: Fairness-Driven Age-Friendly Community Planning via Conditional Diffusion Generation

    cs.AI 2024-12 conditional novelty 6.0 of 10

    A conditional graph diffusion model with a fairness-demand module generates age-friendly facility layouts that outperform baseline planners across five metrics in Beijing.

  23. Pharmacophore-guided de novo drug design with diffusion bridge

    q-bio.BM 2024-12 conditional novelty 6.0 of 10

    PharmacoBridge, an SE(3)-equivariant diffusion bridge, generates valid 3D drug-like molecules directly from pharmacophore point clouds and outperforms pocket-based baselines in pharmacophore matching and docking affin...

  24. Generative Adversarial Reviews: When LLMs Become the Critic

    cs.CL 2024-12 conditional novelty 6.0 of 10

    A new LLM-agent framework, GAR, generates peer reviews from a graph representation of manuscripts and predicts conference acceptance decisions, reportedly matching or exceeding human reviewer performance.

  25. P2DFlow: A Protein Ensemble Generative Model with SE(3) Flow Matching

    physics.bio-ph 2024-11 conditional novelty 6.0 of 10

    A flow-matching model conditioned on a data-derived 'approximate energy' generates protein ensembles matching MD simulations better than AlphaFlow and STR2STR on ATLAS.

  26. Vertical Validation: Evaluating Implicit Generative Models for Graphs on Thin Support Regions

    cs.LG 2024-11 conditional novelty 6.0 of 10

    A biased-splitting plus reweighting evaluation method, Vertical Validation, tests whether implicit graph generators can match held-out data in thinly supported regions and selects models that generalize there.

  27. NGTM: Substructure-based Neural Graph Topic Model for Interpretable Graph Generation

    cs.LG 2025-07 conditional novelty 5.0 of 10

    NGTM generates graphs by sampling substructures from learned topic-specific distributions and assembling them, achieving competitive quality with interpretable, controllable topics.

  28. Semantically Consistent Discrete Diffusion for 3D Biological Graph Modeling

    cs.CV 2025-07 conditional novelty 5.0 of 10

    A stochastic label-consistent projector during discrete diffusion sampling enforces semantic edge-label constraints, yielding anatomically valid 3D vessel and airway graphs that improve downstream labeling and support...

  29. A Deep Generative Model for the Simulation of Discrete Karst Networks

    stat.ML 2025-06 conditional novelty 5.0 of 10

    A two-step pipeline of a graph recurrent network and a graph denoising diffusion model learns to generate stochastic discrete karst networks from subgraphs of two mapped cave systems, reproducing most geometric and to...

  30. Generative AI Enabled Robust Sensor Placement in Cyber-Physical Power Systems: A Graph Diffusion Approach

    eess.SY 2025-01 conditional novelty 5.0 of 10

    A graph-diffusion reinforcement-learning algorithm (EFGD) is proposed to place sensors in a cyber-physical power system while jointly optimizing anomaly-detection coverage and communication-network robustness.

  31. TrojFlow: Flow Models are Natural Targets for Trojan Attacks

    cs.CV 2024-12 conditional novelty 5.0 of 10

    TrojFlow fine-tunes a rectified-flow generator so specific trigger-noise inputs produce attacker-chosen images, keeps benign outputs usable, and qualitatively evades UFID and TERD defenses on CIFAR-10 and CelebA.

  32. Simple Guidance Mechanisms for Discrete Diffusion Models

    cs.LG 2024-12 conditional novelty 5.0 of 10

    Uniform-noise discrete diffusion trained with a continuous-time variational bound (UDLM) plus discrete classifier-free and classifier-based guidance improves controllable generation over autoregressive baselines on ge...

  33. Graph Diffusion-Based AeBS Deployment and Resource Allocation in RSMA-Enabled URLLC Low-Altitude Wireless Networks

    cs.NI 2025-07 conditional novelty 4.0 of 10

    A graph diffusion model and successive convex approximation are alternated to solve joint aerial base station placement, user association, and RSMA resource allocation, with simulated gains over DRL and NOMA baselines.

  34. Non-exchangeable Conformal Prediction for Temporal Graph Neural Networks

    cs.LG 2025-07 reject novelty 4.0 of 10

    NCPNet applies non-exchangeable conformal prediction to temporal graphs by diffusing non-conformity scores over graph and time neighbors and learning weighted quantiles to reduce prediction set size.

  35. Graph Neural Networks in Modern AI-aided Drug Discovery

    q-bio.BM 2025-06 conditional novelty 1.0 of 10

    A comprehensive model-centric review of graph neural network methods and applications in AI-aided drug discovery, from molecular representation to synthesis planning.

Pith tools