Foundation models hold promise for specialized medical imaging tasks, though their effectiveness in breast imaging remains underexplored. This study leverages BiomedCLIP as a foundation model to address challenges in model generalizations. BiomedCLIP was adapted for automated BI-RADS breast density classification using multi-modality mammographic data (synthesized 2D images, digital mammography, and digital breast tomosyn-thesis). Using 96,995 images, we compared single-modality (s2D only) and multi-modality training approaches, addressing class imbalance through weighted contrastive learning. Both approaches achieved similar accuracy (multi-modality: 0.74, single-modality: 0.73), with the multi-modality model offering broader applicability across different imaging modalities and higher AUC values consistently above 0.84 across BI-RADS categories. External validation on the RSNA and EMBED datasets showed strong generalization capabilities (AUC range: 0.80-0.93). GradCAM visualizations confirmed consistent and clinically relevant attention patterns, highlighting the models' interpretability and robustness. This research underscores the potential of foundation models for breast imaging applications, paving the way for future extensions for diagnostic tasks.

Artificial intelligence (AI) systems in high-stakes domains raise concerns about proxy discrimination, unfairness, and explainability. Existing audits often fail to reveal why unfairness arises, particularly when rooted in structural bias. We propose a novel framework using formal abductive explanations to explain proxy discrimination in individual AI decisions. Leveraging background knowledge, our method identifies which features act as unjustified proxies for protected attributes, revealing hidden structural biases. Central to our approach is the concept of aptitude, a task-relevant property independent of group membership, with a mapping function aligning individuals of equivalent aptitude across groups to assess fairness substantively. As a proof of concept, we showcase the framework with examples taken from the German credit dataset, demonstrating its applicability in real-world cases.

Artificial intelligence (AI)-native radio access networks (RANs) will serve vertical industries with stringent requirements: smart grids, autonomous vehicles, remote healthcare, industrial automation, etc. To achieve these requirements, modern 5G/6G design increasingly leverage AI for network optimization, but the opacity of AI decisions poses risks in mission-critical domains. These use cases are often delivered via non-public networks (NPNs) or dedicated network slices, where reliability and safety are vital. In this paper, we motivate the need for transparent and trustworthy AI in high-stakes communications (e.g., healthcare, industrial automation, and robotics) by drawing on 3rd generation partnership project (3GPP)'s vision for non-public networks. We design a mathematical framework to model the trade-offs between transparency (explanation fidelity and fairness), latency, and graphics processing unit (GPU) utilization in deploying explainable AI (XAI) models. Empirical evaluations demonstrate that our proposed hybrid XAI model xAI-Native, consistently surpasses conventional baseline models in performance.

Both humans and machine learning models learn from experience, particularly in safety- and reliability-critical domains. While psychology seeks to understand human cognition, the field of Explainable AI (XAI) develops methods to interpret machine learning models. This study bridges these domains by applying computational tools from XAI to analyze human learning. We modeled human behavior during a complex real-world task -- tuning a particle accelerator -- by constructing graphs of operator subtasks. Applying techniques such as community detection and hierarchical clustering to archival operator data, we reveal how operators decompose the problem into simpler components and how these problem-solving structures evolve with expertise. Our findings illuminate how humans develop efficient strategies in the absence of globally optimal solutions, and demonstrate the utility of XAI-based methods for quantitatively studying human cognition.

Online conversations have become more prevalent on public discussion platforms (e.g. Reddit). With growing controversial topics, it is desirable to summarize not only diverse arguments, but also their rationale and justification. Early studies on text summarization focus on capturing general salient information in source documents, overlooking the argumentative nature of online conversations. Recent research on conversation summarization although considers the argumentative relationship among sentences, fail to explicate deeper argument structure within sentences for summarization. In this paper, we propose a novel task of argument-aware quantitative summarization to reveal the claim-reason structure of arguments in conversations, with quantities measuring argument strength. We further propose ARQUSUMM, a novel framework to address the task. To reveal the underlying argument structure within sentences, ARQUSUMM leverages LLM few-shot learning grounded in the argumentation theory to identify propositions within sentences and their claim-reason relationships. For quantitative summarization, ARQUSUMM employs argument structure-aware clustering algorithms to aggregate arguments and quantify their support. Experiments show that ARQUSUMM outperforms existing conversation and quantitative summarization models and generate summaries representing argument structures that are more helpful to users, of high textual quality and quantification accuracy.

Human explanations of natural language, rationales, form a tool to assess whether models learn a label for the right reasons or rely on dataset-specific shortcuts. Sufficiency is a common metric for estimating the informativeness of rationales, but it provides limited insight into the effects of rationale information on model performance. We address this limitation by relating sufficiency to two modelling paradigms: the ability of models to identify which tokens are part of the rationale (through token classification) and the ability of improving model performance by incorporating rationales in the input (through attention regularisation). We find that highly informative rationales are not likely to help classify the instance correctly. Sufficiency conversely captures the classification impact of the non-rationalised context, which interferes with rationale information in the same input. We also find that incorporating rationale information in model inputs can boost cross-domain classification, but results are inconsistent per task and model type. Finally, sufficiency and token classification appear to be unrelated. These results exemplify the complexity of rationales, showing that metrics capable of systematically capturing this type of information merit further investigation.

Machine learning models that operate on graph-structured data, such as molecular graphs or social networks, often make accurate predictions but offer little insight into why certain predictions are made. Counterfactual explanations address this challenge by seeking the closest alternative scenario where the model's prediction would change. Although counterfactual explanations are extensively studied in tabular data and computer vision, the graph domain remains comparatively underexplored. Constructing graph counterfactuals is intrinsically difficult because graphs are discrete and non-euclidean objects. We introduce Graph Diffusion Counterfactual Explanation, a novel framework for generating counterfactual explanations on graph data, combining discrete diffusion models and classifier-free guidance. We empirically demonstrate that our method reliably generates in-distribution as well as minimally structurally different counterfactuals for both discrete classification targets and continuous properties.

Concept-based explainable artificial intelligence (C-XAI) can let people see which representations an AI model has learned. This is particularly important when high-level semantic information (e.g., actions and relations) is used to make decisions about abstract categories (e.g., danger). In such tasks, AI models need to generalise beyond situation-specific details, and this ability can be reflected in C-XAI outputs that randomise over irrelevant features. However, it is unclear whether people appreciate such generalisation and can distinguish it from other, less desirable forms of imprecision in C-XAI outputs. Therefore, the present study investigated how the generality and relevance of C-XAI outputs affect people's evaluation of AI. In an experimental railway safety evaluation scenario, participants rated the performance of a simulated AI that classified traffic scenes involving people as dangerous or not. These classification decisions were explained via concepts in the form of similar image snippets. The latter differed in their match with the classified image, either regarding a highly relevant feature (i.e., people's relation to tracks) or a less relevant feature (i.e., people's action). Contrary to the hypotheses, concepts that generalised over less relevant features were rated lower than concepts that matched the classified image precisely. Moreover, their ratings were no better than those for systematic misrepresentations of the less relevant feature. Conversely, participants were highly sensitive to imprecisions in relevant features. These findings cast doubts on the assumption that people can easily infer from C-XAI outputs whether AI models have gained a deeper understanding of complex situations.

Artificial intelligence (AI) is increasingly embedded in NHS workflows, but its probabilistic and adaptive behaviour conflicts with the deterministic assumptions underpinning existing clinical-safety standards. DCB0129 and DCB0160 provide strong governance for conventional software yet do not define how AI-specific transparency, interpretability, or model drift should be evidenced within Safety Cases, Hazard Logs, or post-market monitoring. This paper proposes an Explainability-Enabled Clinical Safety Framework (ECSF) that integrates explainability into the DCB0129/0160 lifecycle, enabling Clinical Safety Officers to use interpretability outputs as structured safety evidence without altering compliance pathways. A cross-regulatory synthesis mapped DCB clauses to principles from Good Machine Learning Practice, the NHS AI Assurance and T.E.S.T. frameworks, and the EU AI Act. The resulting matrix links regulatory clauses, principles, ECSF checkpoints, and suitable explainability outputs. ECSF introduces five checkpoints: global transparency for hazard identification, case-level interpretability for verification, clinician usability for evaluation, traceable decision pathways for risk control, and longitudinal interpretability monitoring for post-market surveillance. Techniques such as SHAP, LIME, Integrated Gradients, saliency mapping, and attention visualisation are mapped to corresponding DCB artefacts. ECSF reframes explainability as a core element of clinical-safety assurance, bridging deterministic risk governance with the probabilistic behaviour of AI and supporting alignment with GMLP, the EU AI Act, and NHS AI Assurance principles.

Graph Neural Networks (GNNs) excel in graph-based learning tasks, but their complex, non-linear operations often render them as opaque "black boxes". This opacity hinders user trust, complicates debugging, bias detection, and adoption in critical domains requiring explainability. This paper introduces InteractiveGNNExplainer, a visual analytics framework to enhance GNN explainability, focusing on node classification. Our system uniquely integrates coordinated interactive views (dynamic graph layouts, embedding projections, feature inspection, neighborhood analysis) with established post-hoc (GNNExplainer) and intrinsic (GAT attention) explanation techniques. Crucially, it incorporates interactive graph editing, allowing users to perform a "what-if" analysis by perturbing graph structures and observing immediate impacts on GNN predictions and explanations. We detail the system architecture and, through case studies on Cora and CiteSeer datasets, demonstrate how InteractiveGNNExplainer facilitates in-depth misclassification diagnosis, comparative analysis of GCN versus GAT behaviors, and rigorous probing of model sensitivity. These capabilities foster a deeper, multifaceted understanding of GNN predictions, contributing to more transparent, trustworthy, and robust graph analysis.