Research map / Graph learning / Graph self-supervised learning
Modality-free Graph In-context Alignment
Wei Zhuo, Siqiang Luo
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 MF-GIA instead of the second-best baseline | few-shot node classification · Cora 7-way 5-shot | accuracy: 52.64 -> 63.98 | Improved |
| use MF-GIA in-context inference instead of fine-tuned GFT/AutoGFM | graph foundation models · node classification (15 configurations) | accuracy | Improved |
| add domain-conditioned alignment (domain embedder) vs Prodigy | graph ICL · unseen-domain node classification | accuracy: Cora 5-shot 44.29 -> 63.98 | Improved |
| use episodic pretraining and inference instead of supervised pretraining with prototypes (ablation) | MF-GIA · Computers | accuracy: 38.73 -> 41.49 | Improved |
| use episodic pretraining and inference instead of supervised pretraining with prototypes (ablation) | MF-GIA · Physics | accuracy: 75.26 -> 79.12 | Improved |
| add feature alignment, label alignment and DPAA to GraphSAGE+FT (ablation) | MF-GIA · node classification | accuracy | Improved |
| swap GNN backbone (ablation) | MF-GIA · 1-shot node classification | accuracy | No clear effect |
| apply MF-GIA pretrained on node tasks only to WN18RR 10-way edge classification | MF-GIA · WN18RR 10-way | accuracy: 35.12 vs 40.02 (AutoGFM) at 5-shot | Hurt |
| apply MF-GIA to FB15K237 40-way 1-shot | MF-GIA · FB15K237 40-way | accuracy: 57.17 vs 60.79 (GFT) | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Prodigy — PRODIGY: Enabling In-context Learning Over Graphs (NeurIPS 2023) · better on 6 of 6 shared results
- AutoGFM — AutoGFM: Automated Graph Foundation Model with Adaptive Architecture Customization (ICML 2025) · better on 3 of 4 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 Unified Graph Language Model for Multi-Domain Multi-Task Graph Alignment Instruction Tuning (NeurIPS 2026 · similarity 0.90)
- UniGTE: Unified Graph–Text Encoding for Zero-Shot Generalization across Graph Tasks and Domains (NeurIPS 2025 · similarity 0.89)
- Multi-View Empowered Structural Graph Wordification for Language Models (AAAI 2025 · similarity 0.89)
- Towards Graph Foundation Models: Learning Generalities Across Graphs via Task-Trees (ICML 2025 · similarity 0.88)
- Graph Your Own Prompt (NeurIPS 2025 · similarity 0.87)
- GRAVER: Generative Graph Vocabularies for Robust Graph Foundation Models Fine-tuning (NeurIPS 2025 · similarity 0.87)
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.
