Brief

Researchers have developed an interpretable closed-loop Intelligent Tutoring System (ITS) designed to enhance public speaking skills through multimodal feedback. The system utilizes an XGBoost model to analyze facial, vocal, textual, and oculomotor features from video segments, providing feedback aligned with a seven-dimensional rating scale. Trained on over 10,000 MOOC video segments, the ITS demonstrated scoring accuracy comparable to expert ratings and led to significant skill improvements in adult learners over a 30-day practice period. AI

A new paper evaluates the Tabular Foundation Model (TabPFN) for motor insurance pricing, comparing it against traditional Generalized Linear Models (GLMs) and XGBoost. The study found that TabPFN did not consistently outperform these established methods. Furthermore, TabPFN demonstrated significantly longer inference times and sensitivity to the size of its in-context training set, suggesting it is not yet a viable replacement for current actuarial practices, especially in data-rich environments. AI

This article details the creation of an enterprise-level platform for fraud detection and credit risk assessment. It outlines a modular system design incorporating graph features, BERT-style embeddings, and XGBoost ensembles for robust scoring. The approach emphasizes production readiness and scalability for financial applications. AI

Researchers have developed an AI-based system to predict construction safety outcomes using natural language processing on incident reports. The updated approach utilizes a larger dataset of over 90,000 reports and incorporates new machine learning models like XGBoost and linear SVM, along with model stacking. This method successfully predicts injury severity, type, body part impacted, and incident type, validating the original approach and significantly advancing the field by improving prediction accuracy for injury severity. AI

Researchers have developed a new framework called Algebraic Machine Learning (AML) that learns through algebraic structure decomposition, bypassing traditional numerical optimization. In evaluations, AML demonstrated competitive performance against established methods like Convolutional Neural Networks (CNNs) and XGBoost on small to medium-sized image and tabular datasets. Notably, AML achieved this without requiring validation or cross-validation, relying instead on a generic algebraic inductive bias rather than modality-specific biases. AI

Researchers have developed a new machine learning framework to predict myocardial ischemia using standard non-contrast CT calcium scoring scans. The model incorporates the Agatston score, eight novel "calcium-omics" features, and patient age, demonstrating significant improvements in predictive performance over traditional methods. This approach could enable more accessible cardiovascular risk stratification by leveraging existing imaging data. AI

Researchers have developed a new method for detecting anomalies in unmanned aerial vehicles (UAVs) by combining quantum machine learning with classical techniques. This approach uses a leakage-free evaluation protocol on the TLM:UAV benchmark to distinguish between physical signals and contextual data. While a standalone quantum model did not consistently outperform classical methods, a hybrid XGBoost and Data Reuploading classifier showed promise by improving accuracy when relying solely on physical signals and achieving the lowest false alarm rate in proxy-free evaluations. AI