Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Ashraf Matrawy

Ashraf Matrawy

2 indexed papers

Recent (6 mo)
2
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

Crypto×2ML×1AI×1

Frequent co-authors

Mohamed elShehaby1×
Saba Pourhanifeh1×
AbdulAziz AbdulGhaffar1×

Research Timeline

2026
Threat Modelling using Domain-Adapted Language Models: Empirical Evaluation and Insights

The paper empirically evaluates domain-adapted and general-purpose LLMs for structured threat modelling (STRIDE on 5G security), finding that domain adaptation and model size do not guarantee reliable performance, and that decoding strategies significantly impact output validity.

A No-Defense Defense Against Gradient-Based Adversarial Attacks on ML-NIDS: Is Less More?

The paper demonstrates that simpler, shallower Deep Neural Network architectures with reduced features and ReLU activations can inherently improve the robustness of ML-NIDS against gradient-based adversarial attacks, often outperforming complex, adversarially trained models.

Highlighted terms show continued research focus across papers

Papers

cs.LGcs.CRRecentMay 18, 2026

A No-Defense Defense Against Gradient-Based Adversarial Attacks on ML-NIDS: Is Less More?

Mohamed elShehaby, Ashraf Matrawy

The paper demonstrates that simpler, shallower Deep Neural Network architectures with reduced features and ReLU activations can inherently improve the robustness of ML-NIDS against gradient-based adve…

View →
cs.CRcs.AIRecentMay 11, 2026

Threat Modelling using Domain-Adapted Language Models: Empirical Evaluation and Insights

Saba Pourhanifeh, AbdulAziz AbdulGhaffar, Ashraf Matrawy

The paper empirically evaluates domain-adapted and general-purpose LLMs for structured threat modelling (STRIDE on 5G security), finding that domain adaptation and model size do not guarantee reliable…

View →