A softmax-weighted model merging method combines multiple LLM-based link stealing attack models into one model that beats existing merging baselines across four graph datasets.
The graph neural network model,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Large Language Models Merging for Enhancing the Link Stealing Attack on Graph Neural Networks
A softmax-weighted model merging method combines multiple LLM-based link stealing attack models into one model that beats existing merging baselines across four graph datasets.