A framework extracts a latent state machine from logs, induces a multi-table relational schema, and uses it as a generative prior to create synthetic data that augments real logs for better anomaly detection.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.MA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
State Machine Guided Multi-Relational Synthetic Data from Logs for Anomaly Detection
A framework extracts a latent state machine from logs, induces a multi-table relational schema, and uses it as a generative prior to create synthetic data that augments real logs for better anomaly detection.