FMFaiza Medjek

Faiza Medjek

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Faiza Medjek has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    RefusalGuard-M: a scalable human–machine framework for multi-turn LLM jailbreak evaluation via semantic refusal manifold modeling

    by Michael Tchuindjang, Nathan Duran, Phil Legg, Faiza Medjek

    Existing multi-turn jailbreak evaluation methods increasingly rely on large language models (LLMs) as automated judges to reduce the cost and scalability limitations of human assessment. However, recent studies show that LLM-based evaluators can diverge from human judgments under adversarial strategies involving subtle linguistic and semantic variations, raising reliability concerns in safety-critical domains such as cybersecurity. To address this challenge, we propose Refusal Manifold Guard (RefusalGuard-M), an open-source semantic evaluation framework that constructs a semantic refusal manifold from human-validated refusal responses for assessing LLM jailbreak interactions, including multi-turn scenarios. RefusalGuard-M uses embedding-based geometric representations to measure deviations from refusal behavior, providing a lightweight, interpretable, and reproducible alternative to LLM-based judging. We evaluate the framework across AdvBench, HarmBench, and CyMulTenSet, covering diverse jailbreak strategies, linguistic transformations, and multi-turn scenarios. Results show that RefusalGuard-M achieves strong agreement with human annotations and comparable recall performance to GPT-based evaluators while adopting a conservative evaluation strategy that prioritizes the detection of harmful outputs. On CyMulTenSet, which evaluates past-tense reformulated multi-turn jailbreaks, RefusalGuard-M achieves up to 0.87 recall, compared with 0.86 for GPT-5 and 0.81 for GPT-4, and reduces inference overhead by up to 3.7 $$\times$$ × relative to embedding-based baselines. These findings demonstrate that semantic refusal representations provide an efficient and scalable approach for jailbreak evaluation, particularly in cybersecurity settings where minimizing missed harmful outputs is critical.

  • Open Access

    Ensemble Learning Approaches for Multi-Class Intrusion Detection Systems for the Internet of Vehicles (IoV): A Comprehensive Survey

    by Manal Alharthi, Faiza Medjek, Djamel Djenouri

    The emergence of the Internet of Vehicles (IoV) has revolutionized intelligent transportation and communication systems. However, IoV presents many complex and ever-changing security challenges and thus requires robust cybersecurity protocols. This paper comprehensively describes and evaluates ensemble learning approaches for multi-class intrusion detection systems in the IoV environment. The study evaluates several approaches, such as stacking, voting, boosting, and bagging. A comprehensive review of the literature spanning 2020 to 2025 reveals important trends and topics that require further investigation and the relative merits of different ensemble approaches. The NSL-KDD, CICIDS2017, and UNSW-NB15 datasets are widely used to evaluate the performance of Ensemble Learning-Based Intrusion Detection Systems (ELIDS). ELIDS evaluation is usually carried out using some popular performance metrics, including Precision, Accuracy, Recall, F1-score, and Area Under Receiver Operating Characteristic Curve (AUC-ROC), which were used to evaluate and measure the effectiveness of different ensemble learning methods. Given the increasing complexity and frequency of cyber threats in IoV environments, ensemble learning methods such as bagging, boosting, and stacking enhance adaptability and robustness. These methods aggregate multiple learners to improve detection rates, reduce false positives, and ensure more resilient intrusion detection models that can evolve alongside emerging attack patterns.


  • Open Access

    A Risk Assessment Analysis to Enhance the Security of OT WAN with SD-WAN

    by Van Joshua Abergos, Faiza Medjek

    This paper introduces a comprehensive risk assessment of various wide area network (WAN) technologies as applied to Operational Technology (OT) infrastructures, thus uncovering which WAN technology is best suited for OT to mitigate the risks of Denial of View (DoV), Denial of Control (DoC), and Denial of Service (DoS). A new risk weight-based evaluation approach is proposed following NIST CSF and ISA/IEC 62443 standard risk scoring (RS). In this approach, RS was modified by introducing new risk metrics, namely, risk (Rn), mitigation (Mm), risk prioritization (WRn), and mitigation prioritization (WMm) to create a specialized probability formula to assess risks on OT WAN infrastructure. The proposed formula has been implemented to automate data analysis and risk scoring across nine WAN technologies. The obtained results demonstrated that software-defined wide area network (SD-WAN) has the best security features that even overshadow its vulnerabilities to perform not just as a WAN solution but as a security solution against DoV, DoC, and DoS. Furthermore, this paper identifies and highlights what to prioritize when designing and assessing an SD-WAN setup. In addition, this paper proposes an SD-WAN-based architecture to reduce DoV, DoC, and DoS risks.