A multi-layer cloud IDS uses per-layer ML detection, learned confidence gates, Chroma memory, and LLM escalation with Q-learning adaptive calibration to cut LLM calls by 58.78% while reporting 88.68% overall accuracy.
Overfitting, underfitting and general model overconfidence and under-performance pitfalls and best practices in machine learning and ai
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Multi-Layer Cloud-IDS Pipeline with LLM and Adaptive Q-Learning Calibration
A multi-layer cloud IDS uses per-layer ML detection, learned confidence gates, Chroma memory, and LLM escalation with Q-learning adaptive calibration to cut LLM calls by 58.78% while reporting 88.68% overall accuracy.