Plausible adversarial perturbations, prediction stability, attack success rates, and defensive strategies

The gap

Evidence profile

Sourced from the inline gaps and future work of the source papers, classified as general, spanning 2 journals.

Supporting evidence — 3 representative gaps

• Agentic Artificial Intelligence for Zero-Day Cyber Threat Detection: An Adaptive Reasoning Approach (2026) · Journal of Artificial Intelligence and Technology · doi

  Future work will examine adversarial transferability, multi-feature perturbation, and adaptive threshold poisoning attacks. Future work will focus on extending the framework with full incremental learning, broader real-time deployment, and validation across more diverse cyber-physical and IoT security domains. Although the proposed framework shows strong performance under cross- dataset distribution shifts, its robustness against adversarial attacks has not yet been evaluated.

  generalinline gaps
  Keywords: future adversarial attacks framework examine transferability multi feature perturbation adaptive threshold poisoning focus extending full
• Robust Ensemble of Selectively Strengthened and Augmented Predictors (2026) · doi

  In this paper, we introduced RESSAP, a robust ensemble framework that combines feature-level selection, data augmentation, and clas- sifier randomization to strengthen classifiers against adversarial evasion attacks. Our experimental evaluation shows that RESSAP improves robustness against adversarial evasion while maintaining strong accuracy on benign inputs. However, we acknowledge that the current evaluation is limited to a synthetically generated dataset, which may not fully capture the complexity of real-world applications. In addition, we have not yet compared RESSAP against alternative robust architectures that are specifically designed for adversarial settings. Future work will therefore focus on evaluating the proposed framework on diverse real-world datasets, conducting more exten- sive comparisons with state-of-the-art robust architectures, and refining the feature selection process to further improve adversarial robustness.

  generalfuture work
  Keywords: adversarial ressap robust against framework feature selection evasion evaluation robustness real world architectures introduced ensemble
• A Framework for Optimising Phishing Websites Detection via Ensemble Learning and Synthetic Minority Oversampling Techniques (2026) · International Journal of Management and Data Analytics · doi

  Future studies should investigate plausible adversarial perturbations, prediction stability, attack success rates, and defensive strategies such as adversarial training and robust feature selection. 6 (1), 2026 313 findings to independent datasets and contemporary phishing environments remains to be established.

  generalinline gaps
  Keywords: adversarial future investigate plausible perturbations prediction stability attack success rates defensive strategies training robust feature

Questions about this gap