Contrastive Message Passing lets GNNs apply similarity-preserving transforms to positive edges and dissimilarity-inducing transforms to negative edges via soft positive semidefinite constraints on weights, yielding gains in low-label high-homophily regimes.
Graphsaint: Graph sampling based inductive learning method
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
dataset 1
citation-polarity summary
representative citing papers
GAT uses static attention where neighbor rankings ignore the query node and thus cannot express some graph problems; GATv2 enables dynamic attention and outperforms GAT on 11 OGB and other benchmarks with equal parameters.
A defense framework detects both subgraph and feature-based graph backdoors by exploiting their lower node-neighborhood feature homophily via neighbor-aware reconstruction loss and robust training.
PRAETORIAN reduces GNN backdoor attack success rate to 0.55% with 0.62% clean accuracy drop by targeting the need for many or highly influential trigger nodes.
TypeBandit allocates a global sampling budget at the node-type level via bandits to supply type summaries as contextual signals for attribute completion, delivering dataset-dependent gains when plugged into standard heterogeneous GNNs like R-GCN and HGT.
ScaleGNN uses communication-free sampling and 4D parallelism to scale mini-batch GNN training to 2048 GPUs, achieving 3.5x speedup over prior state-of-the-art on ogbn-products.
RelaNN associates tuples with learnable embeddings and lifts relational queries to jointly process data and embeddings, enabling declarative implementation of graph neural networks inside database systems.
EUPHORIA is a hybrid framework using meta-learning via graph hypernetworks, physics-biased attention in graph transformers, and residual stability correction for few-shot adaptable robotic assembly planning.
citing papers explorer
-
Learning over Positive and Negative Edges with Contrastive Message Passing
Contrastive Message Passing lets GNNs apply similarity-preserving transforms to positive edges and dissimilarity-inducing transforms to negative edges via soft positive semidefinite constraints on weights, yielding gains in low-label high-homophily regimes.
-
How Attentive are Graph Attention Networks?
GAT uses static attention where neighbor rankings ignore the query node and thus cannot express some graph problems; GATv2 enables dynamic attention and outperforms GAT on 11 OGB and other benchmarks with equal parameters.
-
Universal Graph Backdoor Defense: A Feature-based Homophily Perspective
A defense framework detects both subgraph and feature-based graph backdoors by exploiting their lower node-neighborhood feature homophily via neighbor-aware reconstruction loss and robust training.
-
Trapping Attacker in Dilemma: Examining Internal Correlations and External Influences of Trigger for Defending GNN Backdoors
PRAETORIAN reduces GNN backdoor attack success rate to 0.55% with 0.62% clean accuracy drop by targeting the need for many or highly influential trigger nodes.
-
TypeBandit: Type-Level Context Allocation and Reweighting for Effective Attribute Completion in Heterogeneous Graph Neural Networks
TypeBandit allocates a global sampling budget at the node-type level via bandits to supply type summaries as contextual signals for attribute completion, delivering dataset-dependent gains when plugged into standard heterogeneous GNNs like R-GCN and HGT.
-
Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training
ScaleGNN uses communication-free sampling and 4D parallelism to scale mini-batch GNN training to 2048 GPUs, achieving 3.5x speedup over prior state-of-the-art on ogbn-products.
-
Incorporating Deep Learning Design in Database Queries
RelaNN associates tuples with learnable embeddings and lifts relational queries to jointly process data and embeddings, enabling declarative implementation of graph neural networks inside database systems.
-
EUPHORIA: Efficient Universal Planning via Hybrid Optimization for Robust Industrial Robotic Assembly
EUPHORIA is a hybrid framework using meta-learning via graph hypernetworks, physics-biased attention in graph transformers, and residual stability correction for few-shot adaptable robotic assembly planning.