Research map / Probabilistic and scientific modeling / Variational and generative sampling
Semi-Implicit Graph Variational Auto-Encoders
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 |
|---|---|---|---|
| add neighborhood uncertainty sharing to the flexible posterior (ablation) | SIG-VAE vs. Naive SIG-VAE · Cora, Citeseer, Pubmed | AUC: 93.97 -> 94.37; 94.25 -> 95.90; 96.53 -> 96.73 | Helped |
| add neighborhood uncertainty sharing to the flexible posterior (ablation) | SIG-VAE vs. Naive SIG-VAE · Cora, Citeseer, Pubmed | AP: 93.29 -> 94.41; 93.60 -> 95.46; 96.01 -> 96.67 | Mixed |
| use the Bernoulli-Poisson decoder instead of the inner-product decoder (ablation) | SIG-VAE · USAir | AUC: 97.56 -> 94.52 | Hurt |
| use the Bernoulli-Poisson decoder instead of the inner-product decoder (ablation) | SIG-VAE · Power | AUC: 95.04 -> 96.23 | Helped |
| use the Bernoulli-Poisson decoder instead of the inner-product decoder (ablation) | SIG-VAE · Power | AP: 96.50 -> 97.28 | Helped |
| use the Bernoulli-Poisson decoder instead of the inner-product decoder (ablation) | SIG-VAE · Router | AP: 94.94 -> 96.86 | Helped |
| use the Bernoulli-Poisson decoder instead of the inner-product decoder (ablation) | SIG-VAE · Cora, Citeseer, Pubmed | AUC: 94.37 -> 96.04; 95.90 -> 96.43; 96.73 -> 97.01 | Helped |
| use a flexible non-Gaussian posterior instead of the VGAE Gaussian posterior | SIG-VAE · Power | AUC: 71.20 -> 96.23 | Helped |
| remove node attributes (ablation) | SIG-VAE (IP) · Cora | AUC: 94.37 -> 91.14 | Hurt |
| remove node attributes (ablation) | SIG-VAE (IP) · Citeseer | AUC: 95.90 -> 88.72 | Hurt |
| randomly remove 10, 20, 50 or 70 (%) of edges (ablation) | SIG-VAE · Cora | accuracy: 79.5, 78.7, 75.3 and 60.6 | Mixed |
Closest papers by meaning
- D-VAE: A Variational Autoencoder for Directed Acyclic Graphs (NeurIPS 2019 · similarity 0.90)
- Dirichlet Graph Variational Autoencoder (NeurIPS 2020 · similarity 0.89)
- Graph Positional and Structural Encoder (ICML 2024 · similarity 0.86)
- Community-Aware Variational Autoencoder for Continuous Dynamic Networks (AAAI 2025 · similarity 0.86)
- CSG-ODE: ControlSynth Graph ODE For Modeling Complex Evolution of Dynamic Graphs (ICML 2025 · similarity 0.86)
- TreeVI: Reparameterizable Tree-structured Variational Inference for Instance-level Correlation Capturing (NeurIPS 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.