Research map / Graph learning / Graph self-supervised learning
HOG-Diff: Higher-Order Guided Diffusion for Graph Generation
Yiming Huang, Tolga Birdal
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 cell-based topological guide instead of Gaussian noise guide (ablation) | HOG-Diff · QM9 | FCD: 0.829 -> 0.172 | Improved |
| use cell-based guide instead of noise guide (ablation) | HOG-Diff · QM9 | Validity: 91.52 -> 98.74 | Improved |
| use cell-based guide instead of noise guide (ablation) | HOG-Diff · QM9 | NSPDK: 0.0015 -> 0.0003 | Improved |
| use cell-based guide instead of peripheral guide (ablation) | HOG-Diff · QM9 | FCD: 0.305 -> 0.172 | Improved |
| use cell-based guide instead of peripheral guide (ablation) | HOG-Diff · QM9 | Validity: 97.58 -> 98.74 | Improved |
| use cell-based guide instead of noise guide (ablation) | HOG-Diff · ZINC250k | FCD: 1.665 -> 1.533 | Improved |
| use cell-based guide instead of peripheral guide (ablation) | HOG-Diff · ZINC250k | FCD: 1.541 -> 1.533 | Improved |
| use cell-based guide instead of noise guide (ablation) | HOG-Diff · ZINC250k | Validity: 96.78 -> 98.56 | Improved |
| use cell-based guide instead of peripheral guide (ablation) | HOG-Diff · ZINC250k | NSPDK: 0.002 -> 0.001 | No clear effect |
| coarse-to-fine higher-order guided bridge instead of classical noise diffusion | HOG-Diff · spectrum loss training | convergence speed | Improved |
| HOG-Diff vs DeFoG on curvature filtration | HOG-Diff · ZINC250k | kappa_FR distance: 0.728 -> 0.190 | Improved |
| HOG-Diff vs Cometh on curvature filtration | HOG-Diff · SBM | kappa_FR distance: 2.748 -> 1.453 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DeFoG — DeFoG: Discrete Flow Matching for Graph Generation (ICML 2025) · better on 4 of 4 shared results
- DiGress — DiGress: Discrete Denoising diffusion for graph generation (ICLR 2023) · better on 3 of 4 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 Generation with Diffusion Mixture (ICML 2024 · similarity 0.91)
- Advancing Graph Generation through Beta Diffusion (ICLR 2025 · similarity 0.90)
- Hyperbolic Geometric Latent Diffusion Model for Graph Generation (ICML 2024 · similarity 0.89)
- TopoDiffusionNet: A Topology-aware Diffusion Model (ICLR 2025 · similarity 0.89)
- Pard: Permutation-Invariant Autoregressive Diffusion for Graph Generation (NeurIPS 2024 · similarity 0.89)
- Discrete-state Continuous-time Diffusion for Graph Generation (NeurIPS 2024 · similarity 0.89)
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.
