Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Rim Ben Salem has a total of 25 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    An Exposure-Aware Self-Adaptive Architecture for Personalised Security Interventions: Design and Simulation-Based Evaluation

    by Rim Ben Salem

    Organisations rely on warnings, notifications, microlearning, and simulated exercises to promote secure employee behaviour, yet repeated exposure can reduce responsiveness to a specific intervention, and sustained exposure across interventions can make further automated intervention undesirable. This paper proposes an exposure-aware self-adaptive architecture that separates three runtime decisions: whether an observed behavioural gap requires a response, whether accumulated cross-intervention burden still permits automation or calls for human review, and which eligible intervention is currently most suitable. Candidate selection combines intervention-specific responsiveness, which decreases with exposure and recovers over time, with employee-specific efficacy learned from observed outcomes, so that each decision is traceable to auditable runtime state. The architecture is evaluated through simulation; no organisational or employee data are used. A policy-independent synthetic generator (144 worlds, replicated on 144 independent worlds) compares it with nine baselines, including the Upper Confidence Bound (UCB) policy UCB1, Thompson sampling, discounted UCB, a Linear Upper Confidence Bound (LinUCB) contextual bandit, and round-robin rotation. Where repeated delivery attenuates the specific intervention, the architecture exceeds UCB1 by 0.049 and its own ablation by 0.073 on a synthetic outcome scale, but is level with LinUCB and at most marginally ahead of Thompson sampling (differences of at most 0.005) and slightly trails rotation; without exposure effects, exploitative selectors perform better. The contribution is therefore an auditable organisation of adaptation responsibilities, not a superior selection algorithm.

  • Open Access

    LLMs on the Political Compass: Prompt-Conditioned Positioning and Directional Drift

    by Rim Ben Salem, Antonio Vincent Romeo, Esma Aïmeur

    Ever since Large Language Models (LLMs) became easily accessible to the general public, one of the main concerns has been their potential for influencing users and shaping their worldview through frequent interactions. Some critics state that LLMs are inherently biased, while others strongly believe that they cater to users and affirm their preexisting beliefs. In this paper, we investigate how LLMs express political positions under a standardised political questionnaire and how those expressed positions change under prompt conditioning and framing. We do so by focusing on two core research questions: (1) How do large language models express political positions when evaluated using a standardised political questionnaire under stance-inducing prompts? (2) How susceptible are these expressed positions to controlled prompt- and framing-based manipulations? To answer these questions, we propose a dual‑LLM architecture that separates content generation from stance evaluation. The LLMs subject to test respond to the 62 items of the Political Compass Test (PCT) instrument, while an independent Judge LLM, specifically trained as a model-as-a-judge, evaluates the orientation and strength of each response. We conduct three experimental conditions: (T1) baseline testing with the original PCT items; (T2) context‑injection via prompt manipulation applied before presenting the same items; and (T3) question‑level framing in which the items themselves are prefaced with left‑ or right‑leaning cues. Our results show that, under our experimental protocol, most models exhibit a predominantly left-of-centre baseline positioning. However, this apparent ideological tendency is accompanied by pronounced directional drift under prompting, particularly in response to right-leaning contextual priming, which consistently produces larger shifts than left-leaning priming and can even relocate models into right-leaning quadrants. Some safety‑aligned models (notably Phi-4) display high neutrality and refusals, whereas most others still take clear positions with high commitment rates. The raw results, along with a user-friendly dashboard allowing visualisation and further testing, are provided as supplementary materials. This work contributes to the growing literature on AI alignment, risk assessment, and the measurement of prompt-conditioned political output behaviour under a standardised evaluation protocol.

  • Open Access

    A Decoupled Workflow-Aware Attribute-Based Access Control Model for Just-in-Time Authorization

    by Thomas Le Gall, Rim Ben Salem, Alban Gabillon, Nora Boulahia-Cuppens, Frédéric Cuppens

    Attribute-Based Access Control (ABAC) evaluates authorization decisions per request, based on the current attribute context, and does not natively enforce control-flow order or history-based constraints such as Separation-of-Duty and Binding-of-Duty. As a result, ABAC-compliant request sequences may still violate the underlying business process. This paper introduces the Workflow Policy Agent (WPA), a decoupled workflow-aware extension of ABAC that composes request-level authorization with workflow-enabledness conditions to realize Just-in-Time access control over a single workflow, and extends it to concurrent, multi-instance, and hierarchical sub-workflow execution using Hierarchical Coloured Petri Nets.We formally prove three properties. The resulting unified policy is a conservative extension of ABAC. Every accepted execution trace is simultaneously ABAC-authorized and workflow-compliant. Under the stated well-formedness, progress, fairness, and finite-spawning assumptions, we prove WPA-soundness for workflows constructed from a selected subset of workflow patterns, including hierarchical composition. A prototype implementing the model as an infrastructure-agnostic, policy-as-code-compatible sidecar demonstrates feasibility on a loan-approval case study, including across hierarchical workflow boundaries. Using synthetic workloads up to 10^6 task templates and 10^5 concurrent cases, per-request latency overhead remains in the tens of milliseconds across the tested configurations.

  • Federated Intrusion Detection System Based on Unsupervised Machine Learning

    by Maxime Gourceyraud, Rim Ben Salem, Christopher Neal, Frédéric Cuppens, Nora Boulahia-Cuppens

  • Open Access

    AiOfAi: 4th Workshop on Adverse Impacts and Collateral Effects of Artificial Intelligence Technologies

    by Esma Aïmeur, Rim Ben Salem, Dorsaf Sallami, Julita Vassileva, Nora Boulahia-Cuppens

  • Mining Reliable ABAC Policies: A Specificity and Confidence-Aware Extension of Rhapsody

    by Ludjina Benoit, Rim Ben Salem, Nora Boulahia-Cuppens, Frédéric Cuppens

  • From static to dynamic risk indicators in predicting and detecting insider attacks

    by N’fAmoussa Kounon Nanamou, Rim Ben Salem, Anis Bkakria, Nora Cuppens-Boulahia, Frédéric Cuppens

    Cyber insider threats represent one of the most complex and insidious challenges to modern cybersecurity, as they originate from legitimate users whose behaviors may turn malicious over time. Traditional approaches often fail due to their reliance on static risk indicators, necessitating dynamic modeling of human behavior to capture evolving risks. In this paper, we propose a novel framework for detecting insider threats by continuously and dynamically inferring personality-based risk indicators from employees’ writing data using a publicly accessible large language model (Meta AI’s Llama-3.2). These indicators are modeled as time series and processed through AutoRegressive Integrated Moving Average (ARIMA) models to forecast behavioral deviations. Predicted anomalies are subsequently classified using a hybrid ensemble combining Artificial Neural Networks (ANN) and Random Forest (RF) to distinguish benign variations from genuine insider threats. Our framework identifies behavioral anomalies, provides interpretable detection windows, and achieves the following results on CMU-CERT datasets (r4.2/r5.2): recall (90.0%/86.0%), precision (92.6%/87.7%), ROC-AUC (94.7%/92.6%), MSE (0.231/0.304), MTTD (21.3 days/31.72 days) and a median latency under 230 ms for real-time operation. These results demonstrate significant improvements over baseline static approaches in both detection accuracy and temporal prediction capability. This work advances human-centric and proactive insider threat detection by integrating personality-based risk indicators with predictive modeling, providing a scalable and interpretable solution for real-time dynamic risk assessment in enterprise environments.


  • A Privilege Creep-Aware Role Mining Method for Enhanced Access Control Security

    by Vincent Bittard, Rim Ben Salem, Ahmed Bouzid, Sara Imene Boucetta, Frédéric Cuppens, Nora Cuppens-Boulahia

    Role Mining (RM) extracts Role-Based Access Control (RBAC) structures from user-permission assignments to reduce administrative overhead. However, existing approaches usually make the assumption of clean datasets, while real-world systems suffer from anomalies like privilege creep, the gradual accumulation of unnecessary permissions.The proposed approach aims to detect potential privilege crept users who should be reviewed first, and identify legitimate permissions assignments to be expressed in RBAC, reducing management complexity. It consists of a two-step procedure: clean the User-Permission Assignment matrix (UPA) using a clustering and statistical analysis, then build an RBAC state using a regular role mining algorithm.The proposed approach yields an average of 90% in privilege creep detection accuracy and over 95% privilege creep correction, evaluated on synthetically made datasets. Evaluation on real-world datasets demonstrates an average 4-fold reduction in required roles while maintaining at least 80% UPA coverage.

  • Designing a Framework to Tackle the Multifaceted Intricacies of Insider Threats

    by Rim Ben Salem, Frédéric Cuppens, Nora Boulahia-Cuppens, Hager Khechine

  • Open Access

    Federated Intrusion Detection System Based on Unsupervised Machine Learning

    by Maxime Gourceyraud, Rim Ben Salem, Christopher Neal, Frédéric Cuppens, Nora Boulahia Cuppens

    Recent Intrusion Detection System (IDS) research has increasingly moved towards the adoption of machine learning methods. However, most of these systems rely on supervised learning approaches, necessitating a fully labeled training set. In the realm of network intrusion detection, the requirement for extensive labeling can become impractically burdensome. Moreover, while IDS training could benefit from inter-company knowledge sharing, the sensitive nature of cybersecurity data often precludes such cooperation. To address these challenges, we propose an IDS architecture that utilizes unsupervised learning to reduce the need for labeling. We further facilitate collaborative learning through the implementation of a federated learning framework. To enhance privacy beyond what current federated clustering models offer, we introduce an innovative federated K-means++ initialization technique. Our findings indicate that transitioning from a centralized to a federated setup does not significantly diminish performance.