Research Papers

In this paper, we critically examine the effectiveness of the requirement to conduct a Data Protection Impact Assessment (DPIA) in Article 35 of the General Data Protection Regulation (GDPR) in light of fairness metrics. Through this analysis, we explore the role of the fairness principle as introduced in Article 5(1)(a) and its multifaceted interpretation in the obligation to conduct a DPIA. Our paper argues that although there is a significant theoretical role for the considerations of fairness in the DPIA process, an analysis of the various guidance documents issued by data protection authorities on the obligation to conduct a DPIA reveals that they rarely mention the fairness principle in practice.

Applications of deep learning to synthetic media generation allow the creation of convincing forgeries, called DeepFakes, with limited technical expertise. DeepFake detection is an increasingly active research area. In this paper, we analyze an existing DeepFake detection technique based on head pose estimation, which can be applied when fake images are generated with an autoencoder-based face swap. Existing literature suggests that this method is an effective DeepFake detector, and its motivating principles are attractively simple. With an eye towards using these principles to develop new DeepFake detectors, we conduct a reproducibility study of the existing method. We conclude that its merits are dramatically overstated, despite its celebrated status. By investigating this discrepancy we uncover a number of important and generalizable insights related to facial landmark detection, identity-agnostic head pose estimation, and algorithmic bias in DeepFake detectors. Our results correct the current literature's perception of state of the art performance for DeepFake detection.

Electronic Health Records (EHRs) provide a wealth of information for machine learning algorithms to predict the patient outcome from the data including diagnostic information, vital signals, lab tests, drug administration, and demographic information. Machine learning models can be built, for example, to evaluate patients based on their predicted mortality or morbidity and to predict required resources for efficient resource management in hospitals. In this paper, we demonstrate that an attacker can manipulate the machine learning predictions with EHRs easily and selectively at test time by backdoor attacks with the poisoned training data. Furthermore, the poison we create has statistically similar features to the original data making it hard to detect, and can also attack multiple machine learning models without any knowledge of the models. With less than 5% of the raw EHR data poisoned, we achieve average attack success rates of 97% on mortality prediction tasks with MIMIC-III database against Logistic Regression, Multilayer Perceptron, and Long Short-term Memory models simultaneously.

Activation monitoring, which probes a model's internal states using lightweight classifiers, is an emerging tool for AI safety. However, its worst-case robustness under a misalignment threat model--where a model might learn to actively conceal its internal states--remains untested. Focusing on this threat model, we ask: could a model learn to evade previously unseen activation monitors? Our core contribution is to stress-test the learnability of this behavior. We demonstrate that finetuning can create Neural Chameleons: models capable of zero-shot evading activation monitors. Specifically, we fine-tune an LLM to evade monitors for a set of benign concepts (e.g., languages, HTML) when conditioned on a trigger of the form: "You are being probed for {concept}". We show that this learned mechanism generalizes zero-shot: by substituting {concept} with a safety-relevant term like 'deception', the model successfully evades previously unseen safety monitors. We validate this phenomenon across diverse model families (Llama, Gemma, Qwen), showing that the evasion succeeds even against monitors trained post hoc on the model's frozen weights. This evasion is highly selective, targeting only the specific concept mentioned in the trigger, and having a modest impact on model capabilities on standard benchmarks. Using Gemma-2-9b-it as a case study, a mechanistic analysis reveals this is achieved via a targeted manipulation that moves activations into a low-dimensional subspace. While stronger defenses like monitor ensembles and non-linear classifiers show greater resilience, the model retains a non-trivial evasion capability. Our work provides a proof-of-concept for this failure mode and a tool to evaluate the worst-case robustness of monitoring techniques against misalignment threat models.

Agentic AI Framework for Individuals with Disabilities and Neurodivergence: A Multi-Agent System for Healthy Eating, Daily Routines, and Inclusive Well-Being

Salman Jan, Toqeer Ali Syed, Gohar Ali +3

The paper presents a detailed Agentic Artificial Intelligence (AI) model that would enable people with disabilities and neurodivergence to lead healthier lives and have more regular days. The system will use a multi-layer structure; it will include an Application and Interface Layer, an Agents Layer, and a Data Source Layer to provide adaptive, transparent, and inclusive support. Fundamentally, a hybrid reasoning engine will synchronize four special-purpose agents, which include: a personalized-nutrition-based, called a Meal Planner Agent; an adaptive-scheduling-based, called a Reminder Agent; interactive assistance during grocery shopping and cooking, called a Food Guidance Agent; and a continuous-intake-and-physiological-tracking, called a Monitoring Agent. All the agents interact through a central communicative system called the Blackboard/Event Bus, which allows autonomous interaction and real-time feedback loops with multimedia user interfaces. Privacy-sensitive data sources, including electronic health records (EHRs), nutritional databases, wearable sensors, and smart kitchen Internet of Things, are also included in the framework and placed into a policy-controlled layer, which ensures data safety and compliance with consent. Collaborative care and clinician dashboards allow common supervision, and discussable artificial intelligence (XAI) modules give brief explanations of why a decision was made, making users responsible and reliant. The proposed agentic AI framework is an extension beyond traditional assistive systems since it incorporates inclusiveness, personalization, and accessibility at all levels. It displays the intersection of multi-agent reasoning, multi-modal interfaces, and human-centered design that will enable the development of autonomy, health, and digital equity among people with disabilities and neurodivergence.

Integrating SDN and the IoT enhances network control and flexibility. DL-based AAD systems improve security by enabling real-time threat detection in SDN-IoT networks. However, these systems remain vulnerable to adversarial attacks that manipulate input data or exploit model weaknesses, significantly degrading detection accuracy. Existing research lacks a systematic analysis of adversarial vulnerabilities specific to DL-based AAD systems in SDN-IoT environments. This SoK study introduces a structured adversarial threat model and a comprehensive taxonomy of attacks, categorising them into data, model, and hybrid-level threats. Unlike previous studies, we systematically evaluate white, black, and grey-box attack strategies across popular benchmark datasets. Our findings reveal that adversarial attacks can reduce detection accuracy by up to 48.4%, with Membership Inference causing the most significant drop. C&W and DeepFool achieve high evasion success rates. However, adversarial training enhances robustness, and its high computational overhead limits the real-time deployment of SDN-IoT applications. We propose adaptive countermeasures, including real-time adversarial mitigation, enhanced retraining mechanisms, and explainable AI-driven security frameworks. By integrating structured threat models, this study offers a more comprehensive approach to attack categorisation, impact assessment, and defence evaluation than previous research. Our work highlights critical vulnerabilities in existing DL-based AAD models and provides practical recommendations for improving resilience, interpretability, and computational efficiency. This study serves as a foundational reference for researchers and practitioners seeking to enhance DL-based AAD security in SDN-IoT networks, offering a systematic adversarial threat model and conceptual defence evaluation based on prior empirical studies.

The rapid advancement of large language models has intensified public concerns about the potential misuse. Therefore, it is important to build trustworthy AI-generated text detection systems. Existing methods neglect stylistic modeling and mostly rely on static thresholds, which greatly limits the detection performance. In this paper, we propose the Mixture of Stylistic Experts (MoSEs) framework that enables stylistics-aware uncertainty quantification through conditional threshold estimation. MoSEs contain three core components, namely, the Stylistics Reference Repository (SRR), the Stylistics-Aware Router (SAR), and the Conditional Threshold Estimator (CTE). For input text, SRR can activate the appropriate reference data in SRR and provide them to CTE. Subsequently, CTE jointly models the linguistic statistical properties and semantic features to dynamically determine the optimal threshold. With a discrimination score, MoSEs yields prediction labels with the corresponding confidence level. Our framework achieves an average improvement 11.34% in detection performance compared to baselines. More inspiringly, MoSEs shows a more evident improvement 39.15% in the low-resource case. Our code is available at https://github.com/creator-xi/MoSEs.

Any exploit taking advantage of zero-day is called a zero-day attack. Previous research and social media trends show a massive demand for research in zero-day attack detection. This paper analyzes Machine Learning (ML) and Deep Learning (DL) based approaches to create Intrusion Detection Systems (IDS) and scrutinizing them using Explainable AI (XAI) by training an explainer based on randomly sampled data from the testing set. The focus is on using the KDD99 dataset, which has the most research done among all the datasets for detecting zero-day attacks. The paper aims to synthesize the dataset to have fewer classes for multi-class classification, test ML and DL approaches on pattern recognition, establish the robustness and dependability of the model, and establish the interpretability and scalability of the model. We evaluated the performance of four multilayer perceptron (MLP) trained on the KDD99 dataset, including baseline ML models, weighted ML models, truncated ML models, and weighted truncated ML models. Our results demonstrate that the truncated ML model achieves the highest accuracy (99.62%), precision, and recall, while weighted truncated ML model shows lower accuracy (97.26%) but better class representation (less bias) among all the classes with improved unweighted recall score. We also used Shapely Additive exPlanations (SHAP) to train explainer for our truncated models to check for feature importance among the two weighted and unweighted models.

Improving the safety and reliability of large language models (LLMs) is a crucial aspect of realizing trustworthy AI systems. Although alignment methods aim to suppress harmful content generation, LLMs are often still vulnerable to jailbreaking attacks that employ adversarial inputs that subvert alignment and induce harmful outputs. We propose the Randomized Embedding Smoothing and Token Aggregation (RESTA) defense, which adds random noise to the embedding vectors and performs aggregation during the generation of each output token, with the aim of better preserving semantic information. Our experiments demonstrate that our approach achieves superior robustness versus utility tradeoffs compared to the baseline defenses.

Integrating artificial intelligence (AI) and federated learning (FL) in smart transportation has raised critical issues regarding their responsible use. Ensuring responsible AI is paramount for the stability and sustainability of intelligent transportation systems. Despite its importance, research on the responsible application of AI and FL in this domain remains nascent, with a paucity of in-depth investigations into their confluence. Our study analyzes the roles of FL in smart transportation, as well as the promoting effect of responsible AI on distributed smart transportation. Lastly, we discuss the challenges of developing and implementing responsible FL in smart transportation and propose potential solutions. By integrating responsible AI and federated learning, intelligent transportation systems are expected to achieve a higher degree of intelligence, personalization, safety, and transparency.

This study offers an in-depth analysis of the application and implications of the National Institute of Standards and Technology's AI Risk Management Framework (NIST AI RMF) within the domain of surveillance technologies, particularly facial recognition technology. Given the inherently high-risk and consequential nature of facial recognition systems, our research emphasizes the critical need for a structured approach to risk management in this sector. The paper presents a detailed case study demonstrating the utility of the NIST AI RMF in identifying and mitigating risks that might otherwise remain unnoticed in these technologies. Our primary objective is to develop a comprehensive risk management strategy that advances the practice of responsible AI utilization in feasible, scalable ways. We propose a six-step process tailored to the specific challenges of surveillance technology that aims to produce a more systematic and effective risk management practice. This process emphasizes continual assessment and improvement to facilitate companies in managing AI-related risks more robustly and ensuring ethical and responsible deployment of AI systems. Additionally, our analysis uncovers and discusses critical gaps in the current framework of the NIST AI RMF, particularly concerning its application to surveillance technologies. These insights contribute to the evolving discourse on AI governance and risk management, highlighting areas for future refinement and development in frameworks like the NIST AI RMF.

Genetic programming (GP) has the potential to generate explainable results, especially when used for dimensionality reduction. In this research, we investigate the potential of leveraging eXplainable AI (XAI) and large language models (LLMs) like ChatGPT to improve the interpretability of GP-based non-linear dimensionality reduction. Our study introduces a novel XAI dashboard named GP4NLDR, the first approach to combine state-of-the-art GP with an LLM-powered chatbot to provide comprehensive, user-centred explanations. We showcase the system's ability to provide intuitive and insightful narratives on high-dimensional data reduction processes through case studies. Our study highlights the importance of prompt engineering in eliciting accurate and pertinent responses from LLMs. We also address important considerations around data privacy, hallucinatory outputs, and the rapid advancements in generative AI. Our findings demonstrate its potential in advancing the explainability of GP algorithms. This opens the door for future research into explaining GP models with LLMs.