Awesome Few-Shot Learning on Graphs

This repository provides a curated collection of research papers focused on few-shot learning on graphs. It is derived from our survey paper: A Survey of Few-Shot Learning on Graphs: From Meta-Learning to Pre-Training and Prompting. We will update this list regularly. If you notice any errors or missing papers, please feel free to open an issue or submit a pull request.

Table of Contents

• Awesome Few-Shot Learning on Graphs
  • Table of Contents
  • Few-shot Learning Problems on Graphs
  • Few-shot Learning Techiniques on Graphs
  • Meta-Learning Approaches
    • Structure-Based Enhancement
    • Adaptation-Based Enhancement
  • Pre-Training Approaches
    • Pre-Training Strategies
      • Contrastive Strategies
      • Generative Strategies
    • Parameter-efficient Adaptation
      • Prompting on Text-free Graphs
      • Prompting on Text-attributed Graphs
  • Contributing
  • Citations

Few-shot Learning problems on Graphs

Few-shot Learning Techniques on Graphs

Meta-Learning Approaches

Structure-Based Enhancement

1. Graph Prototypical Networks for Few-shot Learning on Attributed Networks. In CIKM'2020, Paper, Code.
   

2. Adaptive Attentional Network for Few-Shot Knowledge Graph Completion. In EMNLP'2020, Paper, Code.
   

3. HMNet: Hybrid Matching Network for Few-Shot Link Prediction. In DASFAA'2021, Paper.
   

4. Tackling Long-Tailed Relations and Uncommon Entities in Knowledge Graph Completion. In EMNLP'2019, Paper, Code.
   

5. Relative and absolute location embedding for few-shot node classification on graph. In AAAI'2021, Paper, Code. 🌟
   

6. Meta-learning on heterogeneous information networks for cold-start recommendation. In KDD'2020, Paper, Code.
   

7. Graph meta learning via local subgraphs. In NeurIPS'2020, Paper, Code.
   

8. Learning to extrapolate knowledge: Transductive few-shot out-of-graph link prediction. In NeurIPS'2020, Paper, Code.
   

9. Towards locality-aware meta-learning of tail node embeddings on networks. In CIKM'2020, Paper, Code.
   

Adaptation-Based Enhancement

1. Graph few-shot learning via knowledge transfer. In AAAI'2020, Paper.
   

2. Meta-Inductive Node Classification across Graphs. In SIGIR'2021, Paper, Code.
   

3. Prototypical networks for few-shot learning. In NeurIPS'2017, Paper, Code.
   

4. Graph Few-shot Learning with Attribute Matching. In CIKM'2020, Paper.
   

5. Adaptive-Step Graph Meta-Learner for Few-Shot Graph Classification. In CIKM'2020, Paper.
   

6. Few-shot link prediction in dynamic networks. In WSDM'2022, Paper.
   

Pre-Training Approaches

Pre-Training Strategies

Contrastive Strategies

1. Deep Graph Contrastive Representation Learning. Preprint, Paper, Code.
   

2. GCC: Graph Contrastive Coding for Graph Neural Network Pre-Training, In KDD'2020, Paper, Code.
   

3. Graph Contrastive Learning with Augmentations, In NeurIPS'2020, Paper, Code.
   

4. SimGRACE: A Simple Framework for Graph Contrastive Learning without Data Augmentation, In WWW'2022, Paper, Code.
   

5. Self-supervised Graph-level Representation Learning with Local and Global Structure, In ICML'2021, Paper, Code.
   

6. Deep Graph Infomax, In ICLR'2019, Paper, Code.
   

7. InfoGraph: Unsupervised Representation Learning on Graphs, In ICLR'2020, Paper, Code.
   

8. Subgraph Contrast for Scalable Self-Supervised Graph Representation Learning, Preprint, Paper, Code.
   

9. Contrastive Multi-View Representation Learning on Graphs, In ICML'2020, Paper, Code.
   

10. Automated Graph Contrastive Learning, In NeurIPS'2021, Paper, Code.
    

11. Contrastive General Graph Matching with Adaptive Augmentation Sampling, In IJCAI'2024, Paper, Code.
    

12. Bringing Your Own View: Graph Contrastive Learning with Generated Views, In WSDM'2022, Paper, Code.
    

13. Graph Contrastive Learning with Adaptive Augmentation, In WWW'2021, Paper, Code.
    

14. Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning, In KDD'2021, Paper, Code.
    

15. Contrastive Pre-Training of GNNs on Heterogeneous Graphs, In CIKM'2021, Paper, Code.
    

16. Pre-training on Large-scale Heterogeneous Graph, In KDD'2021, Paper, Code.
    

17. A Self-supervised Riemannian GNN with Time Varying Curvature for Temporal Graph Learning, In CIKM'2022, Paper.
    

18. Self-supervised Representation Learning on Dynamic Graphs, In CIKM'2021, Paper.
    

19. CPDG: A Contrastive Pre-training Method for Dynamic Graph Neural Networks, In ICDE'2024, Paper, Code.
    

20. Protein representation learning by geometric structure pretraining, In ICLR'2023, Paper, Code.
    

Generative Strategies

1. Variational Graph Auto-Encoders, In ICLR'2016, Paper, Code.
   

2. GPT-GNN: Generative Pre-Training of Graph Neural Networks, In KDD'2020, Paper, Code.
   

3. What's Behind the Mask: Understanding Masked Graph Modeling for Graph Autoencoders, In KDD'2023, Paper, Code.
   

4. Graph Auto-encoder via Neighborhood Wasserst Reconstruction, In ICLR'2022, Paper, Code.
   

5. Self-supervised Representation Learning via Latent Graph Prediction, In NeurIPS'2022, Paper.
   

6. GraphMAE: Self-Supervised Masked Graph Autoencoders, In KDD'2022, Paper, Code.
   

7. GraphMAE2: A Decoding-Enhanced Masked Self-Supervised Graph Learner, In WWW'2023, Paper, Code.
   

8. Mask and Reason: Pre-Training Knowledge Graph Transformers for Complex Logical Queries, In KDD'2022, Paper, Code.
   

9. Structure Pretraining and Prompt Tuning for Knowledge Graph Transfer, In WWW'2023, Paper, Code.
   

10. Zero-shot Item-based Recommendation via Multi-task Product Knowledge Graph Pre-Training, In CIKM'2023, Paper.
    

11. Pre-training on Dynamic Graph Neural Networks, In Neurocomputing'2022, Paper, Code.
    

12. Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecastingn, In KDD'2022, Paper, Code.
    

13. Pre-training Graph Transformer with Multimodal Side Information for Recommendation, In MM'2021, Paper, Code.
    

14. Multi-task Item-attribute Graph Pre-training for Strict Cold-start Item Recommendation, In RecSys'2023, Paper, Code.
    

Parameter-efficient Adaptation

Prompting on Text-free Graphs

1. GPPT: Graph pre-training and prompt tuning to generalize graph neural networks. In KDD'2022, Paper, Code.
   

2. Voucher Abuse Detection with Prompt-based Fine-tuning on Graph Neural Networks. In CIKM'2023, Paper, Code.
   

3. Graphprompt: Unifying pre-training and downstream tasks for graph neural networks. In WWW'2023, Paper, Code. 🌟
   

4. Motif-based prompt learning for universal cross-domain recommendation. In WSDM'2024, Paper.
   

5. Generalized graph prompt: Toward a unification of pre-training and downstream tasks on graphs. In TKDE'2024, Paper,Code. 🌟
   

6. Non-Homophilic Graph Pre-Training and Prompt Learning. Preprint, Paper.
   

7. Text-Free Multi-domain Graph Pre-training: Toward Graph Foundation Models. Preprint, Paper.
   

8. MultiGPrompt for multi-task pre-training and prompting on graphs. In WWW'2024, Paper, Code.
   

9. HetGPT: Harnessing the power of prompt tuning in pre-trained heterogeneous graph neural networks. In WWW'2024, Paper.
   

10. Universal prompt tuning for graph neural networks. In NeurIPS'2023, Paper, Code.
    

11. Inductive Graph Alignment Prompt: Bridging the Gap between Graph Pre-training and Inductive Fine-tuning From Spectral Perspective. In WWW'2024, Paper.
    

12. Sgl-pt: A strong graph learner with graph prompt tuning. Preprint, Paper.
    

13. HGPrompt: Bridging homogeneous and heterogeneous graphs for few-shot prompt learning. In AAAI'2024, Paper, Code.
    

14. PSP: Pre-training and structure prompt tuning for graph neural networks. In Preprint, Paper, Code.
    

15. ULTRA-DP: Unifying graph pre-training with multi-task graph dual prompt. Preprint, Paper, Code.
    

16. Virtual node tuning for few-shot node classification. In KDD'2023, Paper.
    

17. All in one: Multi-task prompting for graph neural networks. In KDD'2023, Paper, Code. 🌟
    

18. DyGPrompt: Learning Feature and Time Prompts on Dynamic Graphs. Preprint, Paper.
    

19. Prompt learning on temporal interaction graphs. Preprint, Paper.
    

Prompting on Text-attributed Graphs

1. Augmenting low-resource text classification with graph-grounded pre-training and prompting. In SIGIR'2023, Paper, Code. 🌟
   

2. Prompt tuning on graph-augmented low-resource text classification. In TKDE'2024, Paper, Code.
   

3. GraphGPT: Graph instruction tuning for large language models. In SIGIR'2024, Paper, Code.
   

4. Natural language is all a graph needs. In EACL'2024, Paper, Code.
   

5. GIMLET: A unified graph-text model for instruction-based molecule zero-shot learning. In NeurIPS'2023, Paper, Code.
   

6. One for all: Towards training one graph model for all classification tasks. In ICLR'2024, Paper, Code. 🌟
   

7. HiGPT: Heterogeneous graph language model. In KDD'2024, Paper, Code.
   

Contributing

👍 Contributions to this repository are highly encouraged!

If you have any relevant resources to share, please feel free to open an issue or submit a pull request.

Citations

If you find this repository useful, please feel free to cite the following works:

Survey Paper

@article{yu2024few,
  title={Few-Shot Learning on Graphs: from Meta-learning to Pre-training and Prompting},
  author={Yu, Xingtong and Fang, Yuan and Liu, Zemin and Wu, Yuxia and Wen, Zhihao and Bo, Jianyuan and Zhang, Xinming and Hoi, Steven CH},
  journal={arXiv preprint arXiv:2402.01440},
  year={2024}
}

GraphPrompt A Representative Prompt Learning Method on Graphs. One of the Most Influential Papers in WWW'23 by Paper Digest (2023-09 Version).

@inproceedings{liu2023graphprompt,
  title={Graphprompt: Unifying pre-training and downstream tasks for graph neural networks},
  author={Liu, Zemin and Yu, Xingtong and Fang, Yuan and Zhang, Xinming},
  booktitle={WWW},
  pages={417--428},
  year={2023}
}

GraphPrompt+ A Generalized Graph Prompt Method.

@article{yu2023generalized,
  title={Generalized Graph Prompt: Toward a Unification of Pre-Training and Downstream Tasks on Graphs},
  author={Yu, Xingtong and Liu, Zhenghao and Fang, Yuan and Liu, Zemin and Chen, Sihong and Zhang, Xinming},
  journal={IEEE Transactions on Knowledge and Data Engineering},
  year={2024}
}

HGPrompt A Heterogeneous Graph Prompt Method.

@inproceedings{yu2023hgprompt,
  title={HGPROMPT: Bridging Homogeneous and Heterogeneous Graphs for Few-shot Prompt Learning},
  author={Yu, Xingtong and Liu, Zemin and Fang, Yuan and Zhang, Xinming},
  booktitle={AAAI},
  pages={16578--16586},
  year={2024}
}

DyGPrompt A Dynamic Graph Prompt Method.

@article{yu2024dygprompt,
  title={DyGPrompt: Learning Feature and Time Prompts on Dynamic Graphs},
  author={Yu, Xingtong and Liu, Zhenghao and Fang, Yuan and Zhang, Xinming},
  journal={arXiv preprint arXiv:2405.13937},
  year={2024}
}

ProNoG A Non-homophilic Graph Prompt Method.

@article{yu2024non,
  title={Non-Homophilic Graph Pre-Training and Prompt Learning},
  author={Yu, Xingtong and Zhang, Jie and Fang, Yuan and Jiang, Renhe},
  journal={arXiv preprint arXiv:2408.12594},
  year={2024}
}

MultiGPrompt A Multi-task Pre-training and Graph Prompt Method.

@inproceedings{yu2024multigprompt,
  title={MultiGPrompt for Multi-Task Pre-Training and Prompting on Graphs},
  author={Yu, Xingtong and Zhou, Chang and Fang, Yuan and Zhang, Xinming},
  booktitle={WWW},
  pages={515--526},
  year={2024}
}

MDGPT A Multi-domain Pre-training and Graph Prompt Method.

@article{yu2024few,
  title={Few-Shot Learning on Graphs: from Meta-learning to Pre-training and Prompting},
  author={Yu, Xingtong and Fang, Yuan and Liu, Zemin and Wu, Yuxia and Wen, Zhihao and Bo, Jianyuan and Zhang, Xinming and Hoi, Steven CH},
  journal={arXiv preprint arXiv:2402.01440},
  year={2024}
}

Methods for Structure Scarce Problem.

@inproceedings{liu2021tail,
  title={Tail-GNN: Tail-node graph neural networks},
  author={Liu, Zemin and Nguyen, Trung-Kien and Fang, Yuan},
  booktitle={KDD},
  pages={1109--1119},
  year={2021}
}

@inproceedings{lu2020meta,
  title={Meta-learning on heterogeneous information networks for cold-start recommendation},
  author={Lu, Yuanfu and Fang, Yuan and Shi, Chuan},
  booktitle={KDD},
  pages={1563--1573},
  year={2020}
}

@article{liu2023locality,
  title={Locality-aware tail node embeddings on homogeneous and heterogeneous networks},
  author={Liu, Zemin and Fang, Yuan and Zhang, Wentao and Zhang, Xinming and Hoi, Steven CH},
  journal={IEEE TKDE},
  volume={36},
  number={6},
  pages={2517--2532},
  year={2023},
  publisher={IEEE}
}

@inproceedings{liu2020towards,
  title={Towards locality-aware meta-learning of tail node embeddings on networks},
  author={Liu, Zemin and Zhang, Wentao and Fang, Yuan and Zhang, Xinming and Hoi, Steven CH},
  booktitle={CIKM},
  pages={975--984},
  year={2020}
}