Tree-based cybersecurity classifiers exhibit decoupled prediction robustness and SHAP explanation stability under black-box attacks, quantified by a new ESI metric alongside the Robustness Index.
In: IEEE Symposium on Security 28 M
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
-
Beyond Gradient-Based Attacks: Adversarial Robustness and Explainability Stability in Cybersecurity Classifiers
Tree-based cybersecurity classifiers exhibit decoupled prediction robustness and SHAP explanation stability under black-box attacks, quantified by a new ESI metric alongside the Robustness Index.