Span masking is usually justified informally: contiguous masks are said to force phrase-level rather than token-level inference. We quantify that argument and then test it. A boundary-screening theorem shows that the evidence a context carries about...
Postoperative magnetic resonance imaging (MRI) captures anatomical and tissue characteristics relevant to glioma prognosis. We evaluated the contribution of compartment-specific imaging features to clinical and molecular prediction using MU-Glioma-Po...
Mobile mapping system (MMS) panoramas routinely capture the survey vehicle, which occludes a substantial portion of the road surface beneath the camera. Reconstructing this large missing region is challenging because of complex road textures, cast sh...
Risk-Budgeted Antichain Extraction (RBAE) selects disjoint regions from an independently trained tree by composing simultaneous regional error and probability-mass bounds. Its certificate controls accepted error across prespecified audit groups, with...
Clinical evidence arrives in sequence, yet a coherent diagnostician’s belief should depend on what was found, not on the order of finding it. Bayesian belief revision is order-invariant as a theorem, which yields an exact, annotation-free test,...
Updating a TinyML intrusion detector can change its decisions even when an authentic artifact is correctly installed: preprocessing, numerical representation, learned parameters, and the operating threshold jointly define the deployed function. We fo...
Deep learning forecasters of stock index direction often report high accuracy, but results are rarely checked under leakage-free protocols, against naive baselines, or across markets. We compared compact deep networks, classical learners, and naive r...
Large language models (LLMs) increasingly recommend cultural venues and tell visitors why, but such explanations can be plausible without being faithful. In a preregistered counterfactual audit, four commercial LLMs recommended venues in three Greek...
Zero-inflated time series (ZITS), characterised by excessive zeros relative to their standard distributions, pose significant challenges for synthetic data generation due to the need to simultaneously preserve the temporal dependencies, the excess ze...
The rapid growth in mobile users and the increasing adoption of data-intensive applications have created high network demands for low-latency and reliable mobility management in mobile communication networks. However, the traditional handover approac...
World models are widely used for planning and control, but it remains unclear whether they can answer counterfactual questions of the form “what would have happened to this individual under a different action.” Such a question sits at the...
Real-time segmentation of the liver parenchyma is a key requirement for computer-assisted guidance in minimally invasive liver surgery. However, existing methods are developed mainly on public cholecystectomy datasets and are therefore poorly represe...
Long COVID is associated with substantial impairment in health-related quality of life (HRQoL), and social determinants of health may improve the identification of individuals with greater HRQoL burden. However, few prediction models have been extern...
Visual malware recognition classifies byte-plot images of executables and reports near-perfect accuracy, yet it is rarely evaluated the way a deployed detector would be. We present ByteScope, a framework that adds a cosine-margin family classifier, a...
Seized-device chat dumps exceed manual review, and case-specific topic lists are not known in advance. We extend SCOPE (Semantic Coherence Pruning and Extraction), which embeds each message and each of 21 predefined topic labels with all-MiniLM-L12-v...
This paper introduces a diagnostic pipeline for evaluating feature-level privacy risk in machine learning (ML) models, anchored by the Coordinate Dependency Index (CDI)—a unitless, cross-modality metric that is designed to quantify the relative...
This study evaluated whether model complexity adds value when differentiating myocardial infarction with non-obstructive coronary arteries (MINOCA) from coronary slow flow in a small clinical tabular dataset. The retrospective cohort included 107 cor...
Label-efficient fraud detection on drifting transaction graphs has two design axes: the predictor (feature representation and model class) and the acquisition rule that chooses which cases analysts label. Active learning evaluations typically vary on...
Mixture of Experts offers flexible graph reasoning by combining multiple views of a graph through a learned router. However, existing Graph-MoE approaches do not explicitly account for input-dependent expert routing when generating post-hoc explanati...
Feature attributions can describe a fitted predictor while differing from a declared engineering reference. This study develops a reference-aligned audit that keeps predictive performance, attribution agreement, estimator precision, and reference evi...
Student activity recognition (SAR) is becoming increasingly important in smart classroom environments, where automatic visual analysis can be used to analyze the activities of students. However, most of the current state-of-the-art deep learning (DL)...
Skew-probabilistic neural networks (SkewPNN) blend Parzen classification with a skewed kernel and metaheuristic tuning. We adapt them to credit-risk probabilities as CRASK (Credit-Risk Adaptive Skew-normal Kernel): a skew-normal kernel with class- an...
This study investigates the integration of explainable artificial intelligence (XAI) and quantile-based machine learning techniques in dynamic portfolio management under changing market conditions. While machine learning models have demonstrated stro...
Deploying accurate deep learning models for retinal microvascular segmentation on resource-constrained devices requires a trade-off between diagnostic precision and computational efficiency. Furthermore, evaluations based on a single fixed data split...
The reported gains of deep learning for fault detection on the Tennessee Eastman Process (TEP) have not been synthesised. In this systematic review, registered retrospectively in OSF Registries, we searched Scopus and the Web of Science Core Collecti...
Complex scientific datasets require interpretable machine-learning approaches capable of extracting structured information while preserving uncertainty and gradual transitions between data regimes. This work proposes an interpretable fuzzy clustering...
Cross-condition bearing fault diagnosis remains challenging because changes in operating conditions can cause distribution shifts that degrade diagnostic performance. Moreover, domain adaptation is not always beneficial and may introduce negative tra...
Semantic answer caching reduces inference cost by reusing previous responses but may repeatedly serve incorrect answers. Prior research primarily addresses cache matching, verification, and freshness, whereas this study investigates an error-attribut...
Subjective Memory Concern (SMC) represents an important stage for early assessment of Alzheimer’s disease (AD), yet diagnosis remains difficult because structural magnetic resonance imaging (sMRI) alterations are subtle and may overlap with nor...
Alzheimer’s disease (AD) is the most common neurodegenerative disorder, and its pathology accumulates for years before symptoms appear, which places the mild cognitive impairment (MCI) stage at the centre of early detection. Machine learning (M...
Machine learning models are increasingly deployed in high-stakes decision systems where predictive performance alone may provide an insufficient basis for responsible model selection. Although explainable artificial intelligence (XAI) offers techniqu...
Whole-slide images and computed tomography scans are commonly decomposed into patches or slices, whereas diagnostic labels are assigned only at the study level, so that which patches or slices support the diagnosis remains unknown. Multiple-instance...
Despite recent advances in multilingual artificial intelligence, Indigenous languages of the Americas remain underrepresented in natural language processing (NLP) because of limited corpora, annotations, benchmarks, and reusable resources. This syste...
Generative models hold great potential but only if one can trust the evaluation of the data they generate. It is common to visualize and evaluate quality on two-dimensional visualizations of high-dimensional data. We show that many commonly used qual...
Physics-informed neural networks (PINNs) approximate solutions of differential equations by embedding governing equations and boundary conditions into the training objective. Controlled PINN comparisons are difficult because geometry representation,...
Electricity theft is a worldwide challenge that causes significant losses for utility companies, with annual revenue losses estimated at USD 89.3 billion globally. The widespread rollout of Advanced Metering Infrastructure (AMI) and smart meters has...
Deep neural decoders have demonstrated strong performance in electroencephalography (EEG)-based motor-imagery (MI) brain–computer interfaces (BCIs), yet their internal representations remain difficult to interpret, limiting neuroscientific insi...
Context selection for large language models (LLMs) requires distinguishing source informativeness from the performance a fixed model can realize. We develop a framework separating the Blackwell order under optimal decision rules from a model-specific...
This paper proposes a formal framework for analyzing extractability in language models based on the minimum number of attempts required to extract a set of sequences with a given confidence. It unifies previously proposed notions and mitigates the th...
We learn compatible destination and owner views in bounded, transcript-free records. Structural source/type/version addressing selects the first relation; authoritative, requested-attribute filtering reduces ordinary candidates to a mean of 3.28. Lea...
Structural dam safety control is fundamentally based on visual inspections and on the analysis of monitoring data and their comparison with the expected behavior for a given load combination, which in turn is calculated with a predictive model. In re...
Nutritional estimation pipelines built on large language models are entering biomarker research, but free-text meal descriptions have no ground truth, so outputs cannot be validated against a reference standard. What can be established without one is...
Large language models can produce convincing explanations that rely on invented facts, unsupported assumptions, or overlooked alternatives. This paper presents a method for checking whether such explanations are supported by the available evidence. T...
Automated microscopic image analysis is essential for quantitative microbiology, yet manual acquisition and generic recognition models struggle with morphological variability and scene artifacts. We address both ends of this pipeline. First, we prese...
Physics-informed neural networks usually incorporate differential equations primarily through the loss function, while the neural architecture itself is selected independently of the underlying numerical structure. This work develops a scheme-induced...
Retrospective clinical research depends on reliable extraction of structured variables from unstructured medical documentation, but manual chart abstraction is slow and vulnerable to inter-rater variability. We compared zero-shot extraction of clinic...
Background: Multi-agent large language model (LLM) systems increasingly rely on structured interaction, debate, and final model-based evaluation, yet an unsupported causal attribution introduced by one participant may propagate through the interactio...
Model validation is essential for evaluating how reliably a machine learning model performs on unseen data, especially in imbalanced classification where performance estimates can be affected by an underrepresentation of the minority class. Cross-val...
The rapid adoption of multimodal learning systems across high-stakes applications has intensified concerns about fairness and social bias, yet existing research remains scattered across modalities, tasks, protected attributes, mitigation strategies,...
In multi-stage manufacturing, conformity is confirmed only at the final test, after the cost has been incurred. On an automotive assembly line (16,500 workpieces), a regression model per checkpoint predicts the final measurement; a threshold turns it...
