2661 papers
This paper proposes a training-free, interpretable framework that leverages criterion-guided prompting and in-context learning with LLMs to detect Alzheimer's disease from manually transcribed speech, achieving state-of-the-art performance on the ADReSS 2020 benchmark while highlighting a critical dependence on high-quality manual transcriptions.
This paper proposes a multi-expert measurement-aware learning framework that leverages multiple expert annotations and a specialized loss function to simultaneously improve carotid intima-media complex segmentation accuracy and millimeter-scale thickness quantification precision in ultrasound images, effectively mitigating the impact of inter-rater variability.
This paper introduces XGNN-ID, an explainable Graph Neural Network framework that integrates a novel Selective Validation-Weighted Ensemble with NodeImport Gating (SVW-NG) to effectively address class imbalance and enhance model interpretability across various graph datasets.
This study proposes a unified, NLP-based framework for Human Resource Management that integrates recruitment, sentiment analysis, and communication tools to enhance efficiency and decision-making, while acknowledging significant challenges related to data privacy, algorithmic bias, and system scalability.
This paper proposes a novel multimodal deep learning framework that integrates EEG, wearable physiological, and behavioral data through a cross-modal attention system and adaptive personalization to achieve highly accurate, early epileptic seizure prediction with reduced false alarms.
This paper proposes a hybrid MCN-MVT framework that combines Capsule Networks and Vision Transformers to achieve high-accuracy, interpretable early detection of Alzheimer's disease, validated by superior performance metrics and LIME-based explanations on MRI data.
The paper introduces STDN-GEN, an automated system leveraging large-language-model agents and a curated vocabulary to rapidly synthesize auditable, four-level critical-material supply chain networks, significantly improving reproducibility and efficiency compared to manual mapping while achieving high accuracy in identifying components and production dependencies.
XTrust-Chain is a multi-layered intrusion detection framework that combines ensemble classifiers with a calibrated trust metric and a cryptographically secured audit trail to enable high-confidence automated decision-making while ensuring immutable verification of alerts across multiple benchmark datasets.
This paper introduces Closure-Guided Optimization (CGO), a constraint-handling framework that utilizes Feasibility Closure Complexity (FCC) to minimize structural repair costs, demonstrating its effectiveness in scenarios where violation rankings diverge from actual repair difficulty while acknowledging it is not a universal advantage over existing methods.
This paper proposes FOIN, a novel federated multi-organoid intelligence architecture that leverages combinatorial graph modeling, neuromorphic computing, and blockchain to enable secure, energy-efficient, and privacy-preserving collaborative learning among distributed brain organoids, achieving high accuracy while significantly reducing computational costs compared to traditional centralized systems.