Research map / Probabilistic and scientific modeling / Variational and generative sampling
Improving Textual Network Learning with Variational Homophilic Embeddings
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 |
|---|---|---|---|
| Use VHE rather than WANE | Textual network link prediction · Zhihu, 15% training edges | AUC: 58.7 -> 66.8 | Helped |
| Add uncertainty modeling by randomly dropping a small portion of observed edges (ablation) | VHE · Cora, HepTh, and Zhihu link prediction | AUC | Helped |
| Add structure embeddings (ablation) | VHE · Cora and HepTh link prediction | AUC: about 2 points | Helped |
| Add structure embeddings (ablation) | VHE · Zhihu link prediction | AUC | No clear effect |
| Increase the homophily factor λ (ablation) | VHE · Cora, 95% training edges | AUC | Helped |
| Use a small positive edge dropout ratio rather than α=0 (ablation) | VHE · Cora, HepTh, and Zhihu link prediction | AUC | Helped |
| Set edge dropout ratio α to 1.0 (ablation) | VHE · Link prediction | AUC | No clear effect |
Closest papers by meaning
- DGE: Deep Generative Network Embedding Based on Commonality and Individuality (AAAI 2020 · similarity 0.88)
- Dynamic Embedding on Textual Networks via a Gaussian Process (AAAI 2020 · similarity 0.88)
- Graph Representation Learning via Ladder Gamma Variational Autoencoders (AAAI 2020 · similarity 0.88)
- Generalization Principles for Inference over Text-Attributed Graphs with Large Language Models (ICML 2025 · similarity 0.86)
- Repurposing Language Models into Embedding Models: Finding the Compute-Optimal Recipe (NeurIPS 2024 · similarity 0.86)
- LLMs as Zero-shot Graph Learners: Alignment of GNN Representations with LLM Token Embeddings (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.