Fine-tuned AI models learn to name the target of online hate
A new study shows that fine-tuning large language models substantially outperforms prompting and classical baselines for detecting which group online ...
A new study shows that fine-tuning large language models substantially outperforms prompting and classical baselines for detecting which group online ...
A new causal graph neural network filters out camouflaged neighbour signals in transaction graphs, outperforming seven baselines on Bitcoin, YelpChi, ...
Researchers have built an AI framework that converts cheap RGB photos of oil palm leaves into 31-band hyperspectral cubes, using ...
Researchers have released PEAXAI, an open-source R package that reframes data envelopment analysis as a probabilistic classification task and uses ...
Researchers have identified a mathematical probability ceiling in ordinal machine learning models that can silently suppress rare blood cell stages, ...
Researchers have developed an explainable deep learning framework that combines Monte Carlo dropout uncertainty, firefly optimization, and SHAP interpretability to ...
Researchers have unveiled Hydra, an interpretable ensemble deep learning framework that jointly performs feature selection and cell type annotation across ...
Researchers in Beijing have built a 0.27-megabyte transformer model that detects multi-stage APT attacks in industrial IoT telemetry with high ...
A new systematic review and meta-analysis finds AI classifiers achieve high specificity for immature white blood cell subtypes in acute ...
A leakage-free benchmark of ten chronic-disease cohorts finds that no outlier-removal method beats doing nothing, while data-cleaning steps silently delete ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine