arXiv subjects

Carsten Maple

Publications and source records attributed to Carsten Maple.

At least 37 records · Page 2Linked to original sources

arXiv2026

SBOMs into Agentic AIBOMs: Schema Extensions, Agentic Orchestration, and Reproducibility Evaluation

Software supply-chain security requires provenance mechanisms that support reproducibility and vulnerability assessment under dynamic execution conditions. Conventional Software Bills of Materials (SBOMs) provide static dependency inventories but cannot capture runtime behaviour, environment drift, or exploitability context. This paper introduces agentic Artificial Intelligence Bills of Materials (AIBOMs), extending SBOMs into active provenance artefacts through autonomous, policy-constrained reasoning. We present an agentic AIBOM framework based on a multi-agent architecture comprising (i) a baseline environment reconstruction agent (MCP), (ii) a runtime dependency and drift-monitoring agent (A2A), and (iii) a policy-aware vulnerability and VEX reasoning agent (AGNTCY). These agents generate contextual exploitability assertions by combining runtime execution evidence, dependency usage, and environmental mitigations with ISO/IEC 20153:2025 Common Security Advisory Framework (CSAF) v2.0 semantics. Exploitability is expressed via structured VEX assertions rather than enforcement actions. The framework introduces minimal, standards-aligned schema extensions to CycloneDX and SPDX, capturing execution context, dependency evolution, and agent decision provenance while preserving interoperability. Evaluation across heterogeneous analytical workloads demonstrates improved runtime dependency capture, reproducibility fidelity, and stability of vulnerability interpretation compared with established provenance systems, with low computational overhead. Ablation studies confirm that each agent contributes distinct capabilities unavailable through deterministic automation.

cs.CR↗

arXiv2026

OpenExtract: Automated Data Extraction for Systematic Reviews in Health

This study presents OpenExtract, an open-source pipeline for automated data extraction in large-scale systematic literature reviews. The pipeline queries large language models (LLMs) to predict data entries based on relevant sections of scientific articles. To test the efficacy of OpenExtract, we apply it to a systematic literature review in digital health and compare its outputs with those of human researchers. OpenExtract achieves precision and recall scores of > 0.8 in this task, indicating that it can be effective at extracting data automatically and efficiently. OpenExtract: https://github.com/JimAchterbergLUMC/OpenExtract.

cs.IR↗

arXiv2025

FraudTransformer: Time-Aware GPT for Transaction Fraud Detection

Detecting payment fraud in real-world banking streams requires models that can exploit both the order of events and the irregular time gaps between them. We introduce FraudTransformer, a sequence model that augments a vanilla GPT-style architecture with (i) a dedicated time encoder that embeds either absolute timestamps or inter-event values, and (ii) a learned positional encoder that preserves relative order. Experiments on a large industrial dataset -- tens of millions of transactions and auxiliary events -- show that FraudTransformer surpasses four strong classical baselines (Logistic Regression, XGBoost and LightGBM) as well as transformer ablations that omit either the time or positional component. On the held-out test set it delivers the highest AUROC and PRAUC.

cs.LG↗

arXiv2025

Fairness-Constrained Optimization Attack in Federated Learning

Federated learning (FL) is a privacy-preserving machine learning technique that facilitates collaboration among participants across demographics. FL enables model sharing, while restricting the movement of data. Since FL provides participants with independence over their training data, it becomes susceptible to poisoning attacks. Such collaboration also propagates bias among the participants, even unintentionally, due to different data distribution or historical bias present in the data. This paper proposes an intentional fairness attack, where a client maliciously sends a biased model, by increasing the fairness loss while training, even considering homogeneous data distribution. The fairness loss is calculated by solving an optimization problem for fairness metrics such as demographic parity and equalized odds. The attack is insidious and hard to detect, as it maintains global accuracy even after increasing the bias. We evaluate our attack against the state-of-the-art Byzantine-robust and fairness-aware aggregation schemes over different datasets, in various settings. The empirical results demonstrate the attack efficacy by increasing the bias up to 90\%, even in the presence of a single malicious client in the FL system.

cs.LG↗

arXiv2025

Private Federated Multiclass Post-hoc Calibration

Calibrating machine learning models so that predicted probabilities better reflect the true outcome frequencies is crucial for reliable decision-making across many applications. In Federated Learning (FL), the goal is to train a global model on data which is distributed across multiple clients and cannot be centralized due to privacy concerns. FL is applied in key areas such as healthcare and finance where calibration is strongly required, yet federated private calibration has been largely overlooked. This work introduces the integration of post-hoc model calibration techniques within FL. Specifically, we transfer traditional centralized calibration methods such as histogram binning and temperature scaling into federated environments and define new methods to operate them under strong client heterogeneity. We study (1) a federated setting and (2) a user-level Differential Privacy (DP) setting and demonstrate how both federation and DP impacts calibration accuracy. We propose strategies to mitigate degradation commonly observed under heterogeneity and our findings highlight that our federated temperature scaling works best for DP-FL whereas our weighted binning approach is best when DP is not required.

cs.LG↗

arXiv2025

Cybersecurity Issues in Local Energy Markets

Local Energy Markets (LEMs), though pivotal to the energy transition, face growing cybersecurity threats due to their reliance on smart grid communication standards and vulnerable Internet-of-Things (IoT)-enabled devices. This is a critical issue because such vulnerabilities can be exploited to manipulate market operations, compromise participants' privacy, and destabilize power distribution networks. This work maps LEM communication flows to existing standards, highlights potential impacts of key identified vulnerabilities, and simulates cyberattack scenarios on a privacy-preserving LEM model to assess their impacts. Findings reveal how attackers could distort pricing and demand patterns. We finally present recommendations for researchers, industry developers, policymakers, and LEM stakeholders to secure future LEM deployments.

cs.CR↗

arXiv2025

Vision Transformer with Adversarial Indicator Token against Adversarial Attacks in Radio Signal Classifications

The remarkable success of transformers across various fields such as natural language processing and computer vision has paved the way for their applications in automatic modulation classification, a critical component in the communication systems of Internet of Things (IoT) devices. However, it has been observed that transformer-based classification of radio signals is susceptible to subtle yet sophisticated adversarial attacks. To address this issue, we have developed a defensive strategy for transformer-based modulation classification systems to counter such adversarial attacks. In this paper, we propose a novel vision transformer (ViT) architecture by introducing a new concept known as adversarial indicator (AdvI) token to detect adversarial attacks. To the best of our knowledge, this is the first work to propose an AdvI token in ViT to defend against adversarial attacks. Integrating an adversarial training method with a detection mechanism using AdvI token, we combine a training time defense and running time defense in a unified neural network model, which reduces architectural complexity of the system compared to detecting adversarial perturbations using separate models. We investigate into the operational principles of our method by examining the attention mechanism. We show the proposed AdvI token acts as a crucial element within the ViT, influencing attention weights and thereby highlighting regions or features in the input data that are potentially suspicious or anomalous. Through experimental results, we demonstrate that our approach surpasses several competitive methods in handling white-box attack scenarios, including those utilizing the fast gradient method, projected gradient descent attacks and basic iterative method.

cs.LG↗

arXiv2025

Individualised Counterfactual Examples Using Conformal Prediction Intervals

Counterfactual explanations for black-box models aim to pr ovide insight into an algorithmic decision to its recipient. For a binary classification problem an individual counterfactual details which features might be changed for the model to infer the opposite class. High-dimensional feature spaces that are typical of machine learning classification models admit many possible counterfactual examples to a decision, and so it is important to identify additional criteria to select the most useful counterfactuals. In this paper, we explore the idea that the counterfactuals should be maximally informative when considering the knowledge of a specific individual about the underlying classifier. To quantify this information gain we explicitly model the knowledge of the individual, and assess the uncertainty of predictions which the individual makes by the width of a conformal prediction interval. Regions of feature space where the prediction interval is wide correspond to areas where the confidence in decision making is low, and an additional counterfactual example might be more informative to an individual. To explore and evaluate our individualised conformal prediction interval counterfactuals (CPICFs), first we present a synthetic data set on a hypercube which allows us to fully visualise the decision boundary, conformal intervals via three different methods, and resultant CPICFs. Second, in this synthetic data set we explore the impact of a single CPICF on the knowledge of an individual locally around the original query. Finally, in both our synthetic data set and a complex real world dataset with a combination of continuous and discrete variables, we measure the utility of these counterfactuals via data augmentation, testing the performance on a held out set.

stat.ML↗

arXiv2025

Justified Evidence Collection for Argument-based AI Fairness Assurance

It is well recognised that ensuring fair AI systems is a complex sociotechnical challenge, which requires careful deliberation and continuous oversight across all stages of a system's lifecycle, from defining requirements to model deployment and deprovisioning. Dynamic argument-based assurance cases, which present structured arguments supported by evidence, have emerged as a systematic approach to evaluating and mitigating safety risks and hazards in AI-enabled system development and have also been extended to deal with broader normative goals such as fairness and explainability. This paper introduces a systems-engineering-driven framework, supported by software tooling, to operationalise a dynamic approach to argument-based assurance in two stages. In the first stage, during the requirements planning phase, a multi-disciplinary and multi-stakeholder team define goals and claims to be established (and evidenced) by conducting a comprehensive fairness governance process. In the second stage, a continuous monitoring interface gathers evidence from existing artefacts (e.g. metrics from automated tests), such as model, data, and use case documentation, to support these arguments dynamically. The framework's effectiveness is demonstrated through an illustrative case study in finance, with a focus on supporting fairness-related arguments.

cs.HC↗

arXiv2025

SoK: Security of EMV Contactless Payment Systems

The widespread adoption of EMV (Europay, Mastercard, and Visa) contactless payment systems has greatly improved convenience for both users and merchants. However, this growth has also exposed significant security challenges. This SoK provides a comprehensive analysis of security vulnerabilities in EMV contactless payments, particularly within the open-loop systems used by Visa and Mastercard. We categorize attacks into seven attack vectors across three key areas: application selection, cardholder authentication, and transaction authorization. We replicate the attacks on Visa and Mastercard protocols using our experimental platform to determine their practical feasibility and offer insights into the current security landscape of contactless payments. Our study also includes a detailed evaluation of the underlying protocols, along with a comparative analysis of Visa and Mastercard, highlighting vulnerabilities and recommending countermeasures.

cs.CR↗

arXiv2025

The Data Sharing Paradox of Synthetic Data in Healthcare

Synthetic data offers a promising solution to privacy concerns in healthcare by generating useful datasets in a privacy-aware manner. However, although synthetic data is typically developed with the intention of sharing said data, ambiguous reidentification risk assessments often prevent synthetic data from seeing the light of day. One of the main causes is that privacy metrics for synthetic data, which inform on reidentification risks, are not well-aligned with practical requirements and regulations regarding data sharing in healthcare. This article discusses the paradoxical situation where synthetic data is designed for data sharing but is often still restricted. We also discuss how the field should move forward to mitigate this issue.

cs.DB↗

arXiv2025

Representation Engineering for Large-Language Models: Survey and Research Challenges

Large-language models are capable of completing a variety of tasks, but remain unpredictable and intractable. Representation engineering seeks to resolve this problem through a new approach utilizing samples of contrasting inputs to detect and edit high-level representations of concepts such as honesty, harmfulness or power-seeking. We formalize the goals and methods of representation engineering to present a cohesive picture of work in this emerging discipline. We compare it with alternative approaches, such as mechanistic interpretability, prompt-engineering and fine-tuning. We outline risks such as performance decrease, compute time increases and steerability issues. We present a clear agenda for future research to build predictable, dynamic, safe and personalizable LLMs.

cs.AI↗

arXiv2025

Assessing the Trustworthiness of Electronic Identity Management Systems: Framework and Insights from Inception to Deployment

The growing dependence on Electronic Identity Management Systems (EIDS) and recent advancements, such as non-human ID management, require a thorough evaluation of their trustworthiness. Assessing EIDS's trustworthiness ensures security, privacy, and reliability in managing sensitive user information. It safeguards against fraud, unauthorised access, and data breaches, fostering user confidence. Existing frameworks primarily focus on specific dimensions such as security and privacy, often neglecting critical dimensions such as ethics, resilience, robustness, and reliability. This paper introduces an integrated Digital Identity Systems Trustworthiness Assessment Framework (DISTAF) encapsulating these six pillars. It is supported by over 65 mechanisms and over 400 metrics derived from international standards and technical guidelines. By addressing the lifecycle of DIMS from design to deployment, our DISTAF evaluates trustworthiness at granular levels while remaining accessible to diverse stakeholders. We demonstrate the application of DISTAF through a real-world implementation using a Modular Open Source Identity Platform (MOSIP) instance, refining its metrics to simplify trustworthiness assessment. Our approach introduces clustering mechanisms for metrics, hierarchical scoring, and mandatory criteria to ensure robust and consistent evaluations across an EIDS in both the design and operation stages. Furthermore, DISTAF is adaptable to emerging technologies like Self-Sovereign Identity (SSI), integrating privacy-enhancing techniques and ethical considerations to meet modern challenges. The assessment tool developed alongside DISTAF provides a user-centric methodology and a simplified yet effective self-assessment process, enabling system designers and assessors to identify system gaps, improve configurations, and enhance public trust.

cs.CR↗

arXiv2025

Cybersecurity Threats to Power Grid Operations from the Demand-Side Response Ecosystem

This article focuses on cyber security threats from IoT-enabled energy smart appliances (ESAs) such as smart heat pumps, electric vehicle chargers, etc., to power grid operations. It presents an in-depth analysis of the demand side threats, including (i) an overview of the vulnerabilities in ESAs and the wider risk from the demand-side response (DSR) ecosystem, (ii) key factors influencing the attack impact on power grid operations, (iii) measures to improve the cyber-physical resilience of power grids, putting them in the context of ongoing efforts from the industry and regulatory bodies worldwide.

cs.CR↗

arXiv2024

Responsible AI Governance: A Response to UN Interim Report on Governing AI for Humanity

This report presents a comprehensive response to the United Nation's Interim Report on Governing Artificial Intelligence (AI) for Humanity. It emphasizes the transformative potential of AI in achieving the Sustainable Development Goals (SDGs) while acknowledging the need for robust governance to mitigate associated risks. The response highlights opportunities for promoting equitable, secure, and inclusive AI ecosystems, which should be supported by investments in infrastructure and multi-stakeholder collaborations across jurisdictions. It also underscores challenges, including societal inequalities exacerbated by AI, ethical concerns, and environmental impacts. Recommendations advocate for legally binding norms, transparency, and multi-layered data governance models, alongside fostering AI literacy and capacity-building initiatives. Internationally, the report calls for harmonising AI governance frameworks with established laws, human rights standards, and regulatory approaches. The report concludes with actionable principles for fostering responsible AI governance through collaboration among governments, industry, academia, and civil society, ensuring the development of AI aligns with universal human values and the public good.

cs.CY↗

arXiv2024

Towards Robust Federated Analytics via Differentially Private Measurements of Statistical Heterogeneity

Statistical heterogeneity is a measure of how skewed the samples of a dataset are. It is a common problem in the study of differential privacy that the usage of a statistically heterogeneous dataset results in a significant loss of accuracy. In federated scenarios, statistical heterogeneity is more likely to happen, and so the above problem is even more pressing. We explore the three most promising ways to measure statistical heterogeneity and give formulae for their accuracy, while simultaneously incorporating differential privacy. We find the optimum privacy parameters via an analytic mechanism, which incorporates root finding methods. We validate the main theorems and related hypotheses experimentally, and test the robustness of the analytic mechanism to different heterogeneity levels. The analytic mechanism in a distributed setting delivers superior accuracy to all combinations involving the classic mechanism and/or the centralized setting. All measures of statistical heterogeneity do not lose significant accuracy when a heterogeneous sample is used.

cs.LG↗

arXiv2024

Privacy-preserving Fuzzy Name Matching for Sharing Financial Intelligence

Financial institutions rely on data for many operations, including a need to drive efficiency, enhance services and prevent financial crime. Data sharing across an organisation or between institutions can facilitate rapid, evidence-based decision-making, including identifying money laundering and fraud. However, modern data privacy regulations impose restrictions on data sharing. For this reason, privacy-enhancing technologies are being increasingly employed to allow organisations to derive shared intelligence while ensuring regulatory compliance. This paper examines the case in which regulatory restrictions mean a party cannot share data on accounts of interest with another (internal or external) party to determine individuals that hold accounts in both datasets. The names of account holders may be recorded differently in each dataset. We introduce a novel privacy-preserving scheme for fuzzy name matching across institutions, employing fully homomorphic encryption over MinHash signatures. The efficiency of the proposed scheme is enhanced using a clustering mechanism. Our scheme ensures privacy by only revealing the possibility of a potential match to the querying party. The practicality and effectiveness are evaluated using different datasets, and compared against state-of-the-art schemes. It takes around 100 and 1000 seconds to search 1000 names from 10k and 100k names, respectively, meeting the requirements of financial institutions. Furthermore, it exhibits significant performance improvement in reducing communication overhead by 30-300 times.

cs.CR↗

arXiv2024

Representation Noising: A Defence Mechanism Against Harmful Finetuning

Releasing open-source large language models (LLMs) presents a dual-use risk since bad actors can easily fine-tune these models for harmful purposes. Even without the open release of weights, weight stealing and fine-tuning APIs make closed models vulnerable to harmful fine-tuning attacks (HFAs). While safety measures like preventing jailbreaks and improving safety guardrails are important, such measures can easily be reversed through fine-tuning. In this work, we propose Representation Noising (RepNoise), a defence mechanism that operates even when attackers have access to the weights. RepNoise works by removing information about harmful representations such that it is difficult to recover them during fine-tuning. Importantly, our defence is also able to generalize across different subsets of harm that have not been seen during the defence process as long as they are drawn from the same distribution of the attack set. Our method does not degrade the general capability of LLMs and retains the ability to train the model on harmless tasks. We provide empirical evidence that the efficacy of our defence lies in its ``depth'': the degree to which information about harmful representations is removed across all layers of the LLM. We also find areas where RepNoise still remains ineffective and highlight how those limitations can inform future research.

cs.CL↗