CONTINUUM detects APT attacks by training a spatial-temporal graph autoencoder on benign provenance graph snapshots and flagging high KNN-distance embeddings as anomalies, optionally under federated learning with homomorphic encryption.
https://mbernste.github.io/posts/gcn/
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CONTINUUM: Detecting APT Attacks through Spatial-Temporal Graph Neural Networks
CONTINUUM detects APT attacks by training a spatial-temporal graph autoencoder on benign provenance graph snapshots and flagging high KNN-distance embeddings as anomalies, optionally under federated learning with homomorphic encryption.