Graph Contrastive Learning

Oct 13, 2025 by recoveryshake.com 69 views

Latest 10 Papers - March 09, 2025

Graph Foundation Model

Graph foundation models have gained significant attention in recent years due to their ability to learn complex patterns and relationships in graph-structured data. These models have been applied to a wide range of tasks, including node classification, link prediction, and graph generation. In this section, we will discuss the latest papers on graph foundation models.

Title	Date	Comment
LLM as GNN: Graph Vocabulary Learning for Text-Attributed Graph Foundation Models	2025-03-05
Exploring the Potential of Large Language Models as Predictors in Dynamic Text-Attributed Graphs	2025-03-05
IOHunter: Graph Foundation Model to Uncover Online Information Operations	2025-03-03	Accep... Accepted at AAAI 2025
Towards Graph Foundation Models: A Study on the Generalization of Positional and Structural Encodings	2025-03-03	Publi... Published at TMLR (https://openreview.net/forum?id=mSoDRZXsqj)
GraphCLIP: Enhancing Transferability in Graph Foundation Models for Text-Attributed Graphs	2025-02-24	Accepted to WWW'25
Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks	2025-02-20
How Expressive are Knowledge Graph Foundation Models?	2025-02-18
Graph Foundation Models for Recommendation: A Comprehensive Survey	2025-02-17
MiNT: Multi-Network Training for Transfer Learning on Temporal Graphs	2025-02-15	20 pa... 20 pages, 9 figures, preprint version
AnomalyGFM: Graph Foundation Model for Zero/Few-shot Anomaly Detection	2025-02-13	14 pages

Graph Prompt

Graph prompts have emerged as a powerful tool for improving the performance of graph neural networks. By providing a prompt to the model, we can guide it to focus on specific aspects of the graph and improve its accuracy. In this section, we will discuss the latest papers on graph prompts.

Title	Date	Comment
Edge Prompt Tuning for Graph Neural Networks	2025-03-02	Accep... Accepted by ICLR 2025
LLM-Empowered Class Imbalanced Graph Prompt Learning for Online Drug Trafficking Detection	2025-02-28
GraphCLIP: Enhancing Transferability in Graph Foundation Models for Text-Attributed Graphs	2025-02-24	Accepted to WWW'25
CuriousLLM: Elevating Multi-Document Question Answering with LLM-Enhanced Knowledge Graph Reasoning	2025-02-18	Accep... Accepted for publication in NAACL 2025. The official version will be available in the ACL Anthology
CLEAR: Cluster-based Prompt Learning on Heterogeneous Graphs	2025-02-13	accep... accepted by PAKDD 2025
HetGPT: Harnessing the Power of Prompt Tuning in Pre-Trained Heterogeneous Graph Neural Networks	2025-02-04	Publi... Published in The ACM Web Conference 2024 (WWW '24)
DAGPrompT: Pushing the Limits of Graph Prompting with a Distribution-aware Graph Prompt Tuning Approach	2025-01-25	To be... To be published in WWW '25, April 28-May 2, 2025, Sydney, NSW, Australia
RELIEF: Reinforcement Learning Empowered Graph Feature Prompt Tuning	2025-01-17	Accep... Accepted by SIGKDD 2025 (camera-ready version). Due to the space limitation, please refer to the V2 version for more details
SAM-Aware Graph Prompt Reasoning Network for Cross-Domain Few-Shot Segmentation	2024-12-31	AAAI 2025
Kernel-Aware Graph Prompt Learning for Few-Shot Anomaly Detection	2024-12-23	Accep... Accepted to AAAI 2025

Graph Contrastive Learning

Graph contrastive learning has emerged as a powerful tool for learning graph representations. By contrasting positive and negative pairs of nodes, we can learn robust and generalizable representations. In this section, we will discuss the latest papers on graph contrastive learning.

Title	Date	Comment
Joint Masked Reconstruction and Contrastive Learning for Mining Interactions Between Proteins	2025-03-06	Submitted
Mixed Graph Contrastive Network for Semi-Supervised Node Classification	2025-03-06
A Binary Classification Social Network Dataset for Graph Machine Learning	2025-03-04
A Fused Gromov-Wasserstein Approach to Subgraph Contrastive Learning	2025-02-28
HyperGCL: Multi-Modal Graph Contrastive Learning via Learnable Hypergraph Views	2025-02-26	9 pages, 2 figures
Understanding and Mitigating Hyperbolic Dimensional Collapse in Graph Contrastive Learning	2025-02-22	This ... This paper is accepted by The 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) 2025
AS-GCL: Asymmetric Spectral Augmentation on Graph Contrastive Learning	2025-02-19	Accepted by TMM
Self-Supervised Graph Contrastive Pretraining for Device-level Integrated Circuits	2025-02-13
Graph Contrastive Learning for Connectome Classification	2025-02-07	Submi... Submitted to EMBC '25
Contrastive Learning Meets Pseudo-label-assisted Mixup Augmentation: A Comprehensive Graph Representation Framework from Local to Global	2025-01-30	9 pages, 2 figures

**
Q&A: Graph Foundation Models, Graph Prompts, and Graph Contrastive Learning

In this article, we will answer some of the most frequently asked questions about graph foundation models, graph prompts, and graph contrastive learning.

Q: What is a graph foundation model?

A: A graph foundation model is a type of machine learning model that is designed to learn complex patterns and relationships in graph-structured data. These models have been applied to a wide range of tasks, including node classification, link prediction, and graph generation.

Q: What is a graph prompt?

A: A graph prompt is a type of input that is used to guide the behavior of a graph neural network. By providing a prompt to the model, we can improve its accuracy and focus its attention on specific aspects of the graph.

Q: What is graph contrastive learning?

A: Graph contrastive learning is a type of machine learning approach that is designed to learn robust and generalizable representations of graphs. By contrasting positive and negative pairs of nodes, we can learn representations that are invariant to different views of the graph.

Q: What are some of the benefits of graph foundation models?

A: Some of the benefits of graph foundation models include:

Improved accuracy on graph-based tasks
Ability to learn complex patterns and relationships in graph-structured data
Ability to handle large and complex graphs
Ability to learn from limited data

Q: What are some of the challenges of graph foundation models?

A: Some of the challenges of graph foundation models include:

Difficulty in handling large and complex graphs
Difficulty in learning complex patterns and relationships in graph-structured data
Difficulty in handling noisy and incomplete data
Difficulty in evaluating the performance of graph foundation models

Q: What are some of the benefits of graph prompts?

A: Some of the benefits of graph prompts include:

Improved accuracy on graph-based tasks
Ability to guide the behavior of graph neural networks
Ability to focus the attention of graph neural networks on specific aspects of the graph
Ability to improve the robustness of graph neural networks

Q: What are some of the challenges of graph prompts?

A: Some of the challenges of graph prompts include:

Difficulty in designing effective prompts
Difficulty in evaluating the performance of graph prompts
Difficulty in handling noisy and incomplete data
Difficulty in handling large and complex graphs

Q: What are some of the benefits of graph contrastive learning?

A: Some of the benefits of graph contrastive learning include:

Improved accuracy on graph-based tasks
Ability to learn robust and generalizable representations of graphs
Ability to handle large and complex graphs
Ability to learn from limited data

Q: What are some of the challenges of graph contrastive learning?

A: Some of the challenges of graph contrastive learning include:

Difficulty in designing effective contrastive learning objectives
Difficulty in evaluating the performance of graph contrastive learning models
Difficulty in handling noisy and incomplete data
Difficulty in handling large and complex graphs

Q: What are some of the applications of graph foundation models, graph prompts, and graph contrastive learning?

A: Some of the applications of graph foundation models, graph prompts, and graph contrastive learning include:

Social network analysis
Recommendation systems
Knowledge graph completion
Graph-based anomaly detection
Graph-based clustering

Q: What are some of the future directions of graph foundation models, graph prompts, and graph contrastive learning?

A: Some of the future directions of graph foundation models, graph prompts, and graph contrastive learning include:

Developing more effective graph foundation models
Developing more effective graph prompts
Developing more effective graph contrastive learning objectives
Developing more effective evaluation metrics for graph foundation models, graph prompts, and graph contrastive learning
Exploring new applications of graph foundation models, graph prompts, and graph contrastive learning.