Research map / Graph learning / Graph self-supervised learning
GraphDF: A Discrete Flow Model for Molecular Graph Generation
Youzhi Luo, Keqiang Yan, Shuiwang Ji
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use discrete latent variables and modulo-shift flows instead of continuous flows | GraphDF molecular generation · ZINC250K | Validity w/o check: 68% -> 89.03% | Improved |
| Use discrete latent variables and modulo-shift flows instead of continuous flows | GraphDF molecular generation · MOSES | Validity w/o check: 71% -> 87.58% | Improved |
| Use discrete latent variables and modulo-shift flows instead of continuous flows | GraphDF property optimization · ZINC250K | Penalized logP, top-1: 12.23 -> 13.7 | Improved |
| Use GraphDF for constrained optimization | GraphDF · JT-VAE/GCPN input set, δ=0.4 | Improvement: 2.49±1.30 -> 4.13±1.41 | Improved |
| Use GraphDF for constrained optimization | GraphDF · GraphAF/MoFlow input set, δ=0.4 | Improvement: 8.21±6.51 -> 9.19±6.43 | Improved |
| Use GraphDF for constrained optimization | GraphDF · GraphAF/MoFlow input set, δ=0.6 | Improvement: 4.98±6.49 -> 4.51±5.80 | Hurt |
| Use GraphDF rather than GraphAF | Generic graph generation · COMMUNITY-SMALL and EGO-SMALL | MMD | Mixed |
Later papers that compared with this method
6 later accepted papers report GraphDF as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- SNACK: A Sequential Notation Framework for Probabilistic Graph Generation (NeurIPS 2026 · better on 1 of 1 shared result)
- Toward a Unified Geometry Understanding : Riemannian Diffusion Framework for Graph Generation and Prediction (NeurIPS 2025 · better on 0 of 2 shared results)
- Unifying Generation and Prediction on Graphs with Latent Graph Diffusion (NeurIPS 2024 · better on 0 of 1 shared result)
- Fast Graph Generation via Spectral Diffusion (TPAMI 2023 · better on 2 of 2 shared results)
- Modular Flows: Differential Molecular Generation (NeurIPS 2022 · better on 5 of 6 shared results)
- Molecule Generation by Principal Subgraph Mining and Assembling (NeurIPS 2022 · better on 2 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GraphAF — GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation (ICLR 2020) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Graph Diffusion that can Insert and Delete (NeurIPS 2025 · similarity 0.88)
- Smooth Interpolation for Improved Discrete Graph Generative Models (ICML 2025 · similarity 0.87)
- Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space (ICLR 2025 · similarity 0.87)
- Graph Diffusion Transformers for Multi-Conditional Molecular Generation (NeurIPS 2024 · similarity 0.87)
- DeFoG: Discrete Flow Matching for Graph Generation (ICML 2025 · similarity 0.86)
- Graph Generation with Diffusion Mixture (ICML 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
