Research map / Graph learning / Graph self-supervised learning
Autoregressive Diffusion Model for Graph Generation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| learned data-dependent diffusion ordering (vs random node ordering) (ablation) | GraphArm · Community-small | Deg./Clus./Orbit MMD: Deg 0.085→0.034, Clus 0.129→0.082, Orbit 0.032→0.004 | Helped |
| learned diffusion ordering (vs random ordering) (ablation) | GraphArm · Community-small | cross-cluster node count during generation: 5.6 → 2.6 | Helped |
| degree-checking procedure during generation | GraphArm · Caveman constrained (max degree 6) | validity: 100% vs 33-39% for one-shot baselines | Helped |
| autoregressive node-absorbing diffusion instead of one-shot diffusion | GraphArm · QM9 | validity: 90.25 vs DiGress 99.0 | Hurt |
| autoregressive node-absorbing diffusion | GraphArm · ZINC250k | FCD: 16.26 vs GDSS 14.66 (GDSS best) | Mixed |
| autoregressive node-absorbing diffusion | GraphArm · Community-small/Caveman generic graphs | generation time: 1.9 s vs GDSS 4.3e2 s on Community-small | Helped |
Closest papers by meaning
- Pard: Permutation-Invariant Autoregressive Diffusion for Graph Generation (NeurIPS 2024 · similarity 0.93)
- Graph Generation with Diffusion Mixture (ICML 2024 · similarity 0.93)
- Discrete-state Continuous-time Diffusion for Graph Generation (NeurIPS 2024 · similarity 0.91)
- SBGD: Improving Graph Diffusion Generative Model via Stochastic Block Diffusion (ICML 2025 · similarity 0.90)
- Generative Modelling of Structurally Constrained Graphs (NeurIPS 2024 · similarity 0.90)
- Generating Graphs via Spectral Diffusion (ICLR 2025 · similarity 0.90)
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.