Research map / Graph learning / GNN architectures
Attribute-guided Dynamic Prompt Learning for Graph Neural Networks
Zhuomin Liang, Liang Bai, Xian Yang
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 |
|---|---|---|---|
| add prompt adaptation module (dynamic prompts) to fixed prompts (ablation) | AttrPrompt · Cora, add 50% edges | accuracy (%): 77.49 -> 77.90 | Improved |
| add prompt adaptation module to fixed prompts (ablation) | AttrPrompt · A-Comp., add 75% edges | accuracy (%): 72.65 -> 73.72 | Improved |
| add prompt adaptation module to fixed prompts (ablation) | AttrPrompt · A-Comp., delete 50% edges | accuracy (%): 79.98 -> 81.10 | Improved |
| add prompt adaptation module to fixed prompts (ablation) | AttrPrompt · Citeseer, delete 50% edges | accuracy (%): 70.71 -> 71.64 | Improved |
| add mutual information loss to dynamic prompts (ablation) | AttrPrompt · A-Comp., add 75% edges | accuracy (%): 73.72 -> 74.16 | Improved |
| add mutual information loss to dynamic prompts (ablation) | AttrPrompt · Cora, delete 50% edges | accuracy (%): 78.73 -> 78.92 | Improved |
| add mutual information loss to dynamic prompts (ablation) | AttrPrompt · A-Photo, add 75% edges | accuracy (%): 77.72 -> 77.73 | No clear effect |
| use attribute-guided prompts on frozen GCN | GCN · A-Photo, add 75% edges | accuracy (%): 67.28 -> 77.73 | Improved |
| use attribute-guided prompts on frozen GCN | GCN · Cora, add 50% edges | accuracy (%): 76.36 -> 78.20 | Improved |
| use attribute-guided prompts on frozen GCN | GCN · A-Comp., add 50% edges | accuracy (%): 69.76 -> 78.04 | Improved |
| use AttrPrompt instead of GOOD-AT | GCN · Cora, add 75% edges | accuracy (%): 76.32 -> 74.43 | Hurt |
| use AttrPrompt instead of CaNet | GCN · A-Photo, add 75% edges | accuracy (%): 79.70 -> 77.73 | Hurt |
Closest papers by meaning
- Node-Time Conditional Prompt Learning in Dynamic Graphs (ICLR 2025 · similarity 0.91)
- Edge Prompt Tuning for Graph Neural Networks (ICLR 2025 · similarity 0.90)
- One Prompt Fits All: Universal Graph Adaptation for Pretrained Models (NeurIPS 2025 · similarity 0.90)
- GraphTOP: Graph Topology-Oriented Prompting for Graph Neural Networks (NeurIPS 2025 · similarity 0.90)
- Towards Dynamic Message Passing on Graphs (NeurIPS 2024 · similarity 0.88)
- HeterGP: Bridging Heterogeneity in Graph Neural Networks with Multi-View Prompting (AAAI 2025 · similarity 0.88)
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.
