Research map / Graph learning / Graph self-supervised learning
Conditional Structure Generation through Graph Variational Generative Adversarial Nets
Carl Yang, Peiye Zhuang, Wenhan Shi, Alan Luu, Pan Li
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 spectral embedding instead of Gaussian random node features (ablation) | CondGen · DBLP seen | LCC difference: 6.70 -> 6.00 | Improved |
| use spectral embedding instead of random node features (ablation) | CondGen · DBLP unseen | LCC difference: 9.25 -> 6.33 | Improved |
| use spectral embedding instead of random node features (ablation) | CondGen · TCGA seen | GINI difference: 0.1509 -> 0.1093 | Improved |
| use spectral embedding instead of random node features (ablation) | CondGen · DBLP | training runtime (minutes): 31.5 -> 72.3 | Hurt |
| spectral embedding vs random features overall (ablation) | CondGen · Seen and unseen conditions | graph statistics | Mixed |
| replace GVAE with CondGen (GCN encoder with latent conjugation plus GCN discriminator) | GVAE · DBLP seen | GINI difference: 0.1964 -> 0.0959 | Improved |
| replace GraphRNN with CondGen | GraphRNN · DBLP seen | LCC difference: 6.88 -> 6.00 | Improved |
| replace GraphRNN with CondGen | GraphRNN · TCGA seen | LCC difference: 16.20 -> 23.72 | Hurt |
| replace GraphRNN with CondGen | GraphRNN · DBLP | training runtime (minutes): 299.5 -> 72.3 | Improved |
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.
- GraphRNN — GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models (ICML 2018) · better on 7 of 9 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
- A Systematic Survey on Deep Generative Models for Graph Generation (TPAMI 2022 · similarity 0.89)
- Evaluating Graph Generative Models with Contrastively Learned Features (NeurIPS 2022 · similarity 0.89)
- GRAVER: Generative Graph Vocabularies for Robust Graph Foundation Models Fine-tuning (NeurIPS 2025 · similarity 0.87)
- Generative Graph Pattern Machine (NeurIPS 2025 · similarity 0.86)
- On the Role of Edge Dependency in Graph Generative Models (ICML 2024 · similarity 0.86)
- Graph Generative Pre-trained Transformer (ICML 2025 · 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.
