Research Papers

Tax authorities and public-sector financial agencies rely on large volumes of unstructured and semi-structured fiscal documents - including tax forms, instructions, publications, and jurisdiction-specific guidance - to support compliance analysis and audit workflows. While recent advances in generative AI and retrieval-augmented generation (RAG) have shown promise for document-centric question answering, existing approaches often lack the transparency, citation fidelity, and conservative behaviour required in high-stakes regulatory domains. This paper presents a multimodal, citation-enforced RAG framework for fiscal document intelligence that prioritises explainability and auditability. The framework adopts a source-first ingestion strategy, preserves page-level provenance, enforces citations during generation, and supports abstention when evidence is insufficient. Evaluation on real IRS and state tax documents demonstrates improved citation fidelity, reduced hallucination, and analyst-usable explanations, illustrating a pathway toward trustworthy AI for tax compliance.

Although concept-based models promise interpretability by explaining predictions with human-understandable concepts, they typically rely on exhaustive annotations and treat concepts as flat and independent. To circumvent this, recent work has introduced Hierarchical Concept Embedding Models (HiCEMs) to explicitly model concept relationships, and Concept Splitting to discover sub-concepts using only coarse annotations. However, both HiCEMs and Concept Splitting are restricted to shallow hierarchies. We overcome this limitation with Multi-Level Concept Splitting (MLCS), which discovers multi-level concept hierarchies from only top-level supervision, and Deep-HiCEMs, an architecture that represents these discovered hierarchies and enables interventions at multiple levels of abstraction. Experiments across multiple datasets show that MLCS discovers human-interpretable concepts absent during training and that Deep-HiCEMs maintain high accuracy while supporting test-time concept interventions that can improve task performance.

Why Does It Look There? Structured Explanations for Image Classification

Jiarui Li, Zixiang Yin, Samuel J Landry +2

Deep learning models achieve remarkable predictive performance, yet their black-box nature limits transparency and trustworthiness. Although numerous explainable artificial intelligence (XAI) methods have been proposed, they primarily provide saliency maps or concepts (i.e., unstructured interpretability). Existing approaches often rely on auxiliary models (\eg, GPT, CLIP) to describe model behavior, thereby compromising faithfulness to the original models. We propose Interpretability to Explainability (I2X), a framework that builds structured explanations directly from unstructured interpretability by quantifying progress at selected checkpoints during training using prototypes extracted from post-hoc XAI methods (e.g., GradCAM). I2X answers the question of "why does it look there" by providing a structured view of both intra- and inter-class decision making during training. Experiments on MNIST and CIFAR10 demonstrate effectiveness of I2X to reveal prototype-based inference process of various image classification models. Moreover, we demonstrate that I2X can be used to improve predictions across different model architectures and datasets: we can identify uncertain prototypes recognized by I2X and then use targeted perturbation of samples that allows fine-tuning to ultimately improve accuracy. Thus, I2X not only faithfully explains model behavior but also provides a practical approach to guide optimization toward desired targets.

Do Machines Fail Like Humans? A Human-Centred Out-of-Distribution Spectrum for Mapping Error Alignment

Binxia Xu, Xiaoliang Luo, Luke Dickens +1

Determining whether AI systems process information similarly to humans is central to cognitive science and trustworthy AI. While modern AI models can match human accuracy on standard tasks, such parity does not guarantee that their underlying decision-making strategies resemble those of humans. Assessing performance using error alignment metrics to compare how humans and models fail, and how this changes for distorted, or otherwise more challenging, stimuli, provides a viable pathway toward a finer characterization of model-human alignment. However, existing out-of-distribution (OOD) analyses for challenging stimuli are limited due to methodological choices: they define OOD shift relative to model training data or use arbitrary distortion-specific parameters with little correspondence to human perception, hindering principled comparisons. We propose a human-centred framework that redefines the degree of OOD as a spectrum of human perceptual difficulty. By quantifying how much a collection of stimuli deviates from an undistorted reference set based on human accuracy, we construct an OOD spectrum and identify four distinct regimes of perceptual challenge. This approach enables principled model-human comparisons at calibrated difficulty levels. We apply this framework to object recognition and reveal unique, regime-dependent model-human alignment rankings and profiles across deep learning architectures. Vision-language models are most consistently human aligned across near- and far-OOD conditions, but convolutional neural networks (CNNs) are more aligned than vision transformers (ViTs) for near-OOD and ViTs are more aligned than CNNs for far-OOD. Our work demonstrates the critical importance of accounting for cross-condition differences, such as perceptual difficulty, for a principled assessment of model-human alignment.

Distinguishing legitimate critical speech from coordinated conspiracy narratives is a pressing challenge for the integrity of the online information ecosystem, yet prevailing approaches rely on labeled entity roles and rigid predefined schemas that fail to capture the diffuse and context-dependent nature of participation in such narratives. We propose a modular framework that extracts an entity-interaction graph directly from narratives and reasons over it for classification, without depending on annotated node-level labels. Our pipeline enumerates candidate entity spans from a domain-adapted BERT encoder and learns a bipartite graph structure, which is then classified with a Heterogeneous Graph Transformer. To make predictions interpretable, we cast the graph as a hypergraph and apply a counterfactual distillation procedure that estimates the Individual Treatment Effect of each node, pruning extraneous entities to recover a minimal predictive subgraph. On the standard PAN Oppositional Narratives benchmark, our model achieves a macro F1-score of 0.93 and an MCC of 0.84, reaching state-of-the-art performance with roughly one-third of the parameters of the next best system, while pruning the graphs by 75.2% on average without altering predictions. Our results demonstrate that graph-based representations coupled with counterfactual inference offer both strong predictive accuracy and transparency into the entities driving conspiratorial discourse.

Autonomous offensive agents often fail to transfer beyond the networks on which they are trained. We isolate a minimal but fundamental shift -- unseen host/subnet IP reassignment in an otherwise fixed enterprise scenario -- and evaluate attacker generalization in the NetSecGame environment. Agents are trained on five IP-range variants and tested on a sixth unseen variant; only the meta-learning agent may adapt at test time. We compare three agent families (traditional RL, adaptation agents, and LLM-based agents) and use action-distribution-based behavioral/XAI analyses to localize failure modes. Some adaptation methods show partial transfer but significant degradation under unseen reassignment, indicating that even address-space changes can break long-horizon attack policies. Under our evaluation protocol and agent-specific assumptions, prompt-driven pretrained LLM agents achieve the highest success on the held-out reassignment, but at the cost of increased inference-time compute, reduced transparency, and practical failure modes such as repetition/invalid-action loops.

While explainable AI (XAI) is often heralded as a means to enhance transparency and trustworthiness in closed-loop neurotechnology for psychiatric and neurological conditions, its real-world prevalence remains low. Moreover, empirical evidence suggests that the type of explanations provided by current XAI methods often fails to align with clinicians' end-user needs. In this viewpoint, we argue that clinically meaningful explainability (CME) is essential for AI-enabled closed-loop medical neurotechnology and must be addressed from an ethical, technical, and clinical perspective. Instead of exhaustive technical detail, clinicians prioritize clinically relevant, actionable explanations, such as clear representations of input-output relationships and feature importance. Full technical transparency, although theoretically desirable, often proves irrelevant or even overwhelming in practice, as it may lead to informational overload. Therefore, we advocate for CME in the neurotechnology domain: prioritizing actionable clarity over technical completeness and designing interface visualizations that intuitively map AI outputs and key features into clinically meaningful formats. To this end, we introduce a reference architecture called NeuroXplain, which translates CME into actionable technical design recommendations for any future neurostimulation device. Our aim is to inform stakeholders working in neurotechnology and regulatory framework development to ensure that explainability fulfills the right needs for the right stakeholders and ultimately leads to better patient treatment and care.

Disruptions at critical logistics nodes pose severe risks to global supply chains, yet existing risk prediction systems typically prioritize forecasting accuracy without providing operationally interpretable early warnings. This paper proposes an evidence-grounded framework that jointly performs supply chain bottleneck prediction and faithful natural-language risk explanation by coupling a Temporal Graph Attention Network (TGAT) with a structured large language model (LLM) reasoning module. Using maritime hubs as a representative case study for global supply chain nodes, daily spatial graphs are constructed from Automatic Identification System (AIS) broadcasts, where inter-node interactions are modeled through attention-based message passing. The TGAT predictor captures spatiotemporal risk dynamics, while model-internal evidence -- including feature z-scores and attention-derived neighbor influence -- is transformed into structured prompts that constrain LLM reasoning to verifiable model outputs. To evaluate explanatory reliability, we introduce a directional-consistency validation protocol that quantitatively measures agreement between generated risk narratives and underlying statistical evidence. Experiments on six months of real-world logistics data demonstrate that the proposed framework outperforms baseline models, achieving a test AUC of 0.761, AP of 0.344, and recall of 0.504 under a strict chronological split while producing early warning explanations with 99.6\% directional consistency. Results show that grounding LLM generation in graph-model evidence enables interpretable and auditable risk reporting without sacrificing predictive performance. The framework provides a practical pathway toward operationally deployable explainable AI for supply chain risk early warning and resilience management.

Explainable Artificial Intelligence (XAI) methods (SHAP, LIME) are increasingly adopted to interpret models in high-stakes businesses. However, the credibility of these explanations, their stability under realistic data perturbations, remains unquantified. This paper introduces the Credibility Index via Explanation Stability (CIES), a mathematically grounded metric that measures how robust a model's explanations are when subject to realistic business noise. CIES captures whether the reasons behind a prediction remain consistent, not just the prediction itself. The metric employs a rank-weighted distance function that penalizes instability in the most important features disproportionately, reflecting business semantics where changes in top decision drivers are more consequential than changes in marginal features. We evaluate CIES across three datasets (customer churn, credit risk, employee attrition), four tree-based classification models and two data balancing conditions. Results demonstrate that model complexity impacts explanation credibility, class imbalance treatment via SMOTE affects not only predictive performance but also explanation stability, and CIES provides statistically superior discriminative power compared to a uniform baseline metric (p < 0.01 in all 24 configurations). A sensitivity analysis across four noise levels confirms the robustness of the metric itself. These findings offer business practitioners a deployable "credibility warning system" for AI-driven decision support.

Multiple Sclerosis (MS) is a chronic autoimmune disease of the central nervous system whose molecular mechanisms remain incompletely understood. In this study, we developed an end-to-end machine learning pipeline to analyze transcriptomic data from peripheral blood mononuclear cells and cerebrospinal fluid, integrating both bulk microarray and single-cell RNA sequencing datasets (concentrating on CD4+ and B-cells). After rigorous preprocessing, batch correction, and gene declustering, XGBoost classifiers were trained to distinguish MS patients from healthy controls. Explainable AI tools, namely SHapley Additive exPlanations (SHAP), were employed to identify key genes driving classification, and results were compared with Differential Expression Analysis (DEA). SHAP-prioritized genes were further investigated through interaction networks and pathway enrichment analyses. The models achieved strong performance, particularly in CSF B-cells (AUC=0.94) and microarray (AUC=0.86). SHAP gene selection proved to be complementary to classical DEA. Gene clusters identified across multiple datasets highlighted immune activation, non-canonical immune checkpoints (ITK, CLEC2D, KLRG1, CEACAM1), ribosomal and translational programs, ubiquitin-proteasome regulation, lipid trafficking, and Epstein-Barr virus-related pathways. Our integrative and explainable framework reveals complementary insights beyond conventional analysis and provides novel mechanistic hypotheses and potential biomarkers for MS pathogenesis.

Continuous physiological monitoring is central to emergency care, yet deploying trustworthy AI is challenging. While LLMs can translate complex physiological signals into clinical narratives, it is unclear how agentic systems perform relative to zero-shot inference. To address these questions, we present Vivaldi, a role-structured multi-agent system that explains multivariate physiological time series. Due to regulatory constraints that preclude live deployment, we instantiate Vivaldi in a controlled, clinical pilot to a small, highly qualified cohort of emergency medicine experts, whose evaluations reveal a context-dependent picture that contrasts with prevailing assumptions that agentic reasoning uniformly improves performance. Our experiments show that agentic pipelines substantially benefit non-thinking and medically fine-tuned models, improving expert-rated explanation justification and relevance by +6.9 and +9.7 points, respectively. Contrarily, for thinking models, agentic orchestration often degrades explanation quality, including a 14-point drop in relevance, while improving diagnostic precision (ESI F1 +3.6). We also find that explicit tool-based computation is decisive for codifiable clinical metrics, whereas subjective targets, such as pain scores and length of stay, show limited or inconsistent changes. Expert evaluation further indicates that gains in clinical utility depend on visualization conventions, with medically specialized models achieving the most favorable trade-offs between utility and clarity. Together, these findings show that the value of agentic AI lies in the selective externalization of computation and structure rather than in maximal reasoning complexity, and highlight concrete design trade-offs and learned lessons, broadly applicable to explainable AI in safety-critical healthcare settings.

Explainable AI (XAI) has become essential in computer vision to make the decision-making processes of deep learning models transparent. However, current visual explanation (XAI) methods face a critical trade-off between the high fidelity of architecture-specific methods and the broad applicability of universal ones. This often results in abstract or fragmented explanations and makes it difficult to compare explanatory power across diverse model families, such as CNNs and Transformers. This paper introduces the Self-Confidence and Analysis Networks (SCAN), a novel universal framework that overcomes these limitations for both convolutional neural network and transformer architectures. SCAN utilizes an AutoEncoder-based approach to reconstruct features from a model's intermediate layers. Guided by the Information Bottleneck principle, it generates a high-resolution Self-Confidence Map that identifies information-rich regions. Extensive experiments on diverse architectures and datasets demonstrate that SCAN consistently achieves outstanding performance on various quantitative metrics such as AUC-D, Negative AUC, Drop%, and Win%. Qualitatively, it produces significantly clearer, object-focused explanations than existing methods. By providing a unified framework that is both architecturally universal and highly faithful, SCAN enhances model transparency and offers a more reliable tool for understanding the decision-making processes of complex neural networks.