PromptCGL learns per-task prompts on a frozen graph neural network and achieves near-joint-training accuracy on four continual graph learning benchmarks with constant memory and near-zero forgetting.
Overcoming catastrophic forgetting in graph neural networks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Prompt-Driven Continual Graph Learning
PromptCGL learns per-task prompts on a frozen graph neural network and achieves near-joint-training accuracy on four continual graph learning benchmarks with constant memory and near-zero forgetting.