3020 papers
This paper proposes a risk-oriented verification and validation framework that transforms hallucination detection in public-sector LLMs from a purely technical challenge into a governance issue by introducing a composite risk index to guide operational decisions like deferral or human escalation, thereby ensuring legal certainty and administrative accountability.
This paper introduces a causal, token-level survival analysis framework that successfully forecasts the onset of hallucinations in large language models before they occur, achieving a 0.777 AUROC by detecting text-novelty drift and providing warnings approximately 11 tokens earlier than traditional post-hoc detectors.
This paper proposes NPDO, a hybrid neural framework combining multi-scale feature extraction with hierarchical CNN and Vision Transformer architectures to intelligently prune the VVC intra-prediction search space, achieving a 54.0% reduction in encoding time with only a 1.18% BD-Rate increase compared to the VTM 23.0 reference.
This paper proposes ADONIS, a decentralized framework for 6G network slicing that leverages a hierarchical edge-federated Digital Twin architecture, a Spatio-Temporal Graph Neural Network scheduler, and a novel delta-encoding protocol to minimize synchronization overhead and SLA violations while ensuring resource allocation convergence.
This paper proposes the Focus-Aware Adaptive Dense Subgraph (FAADS) framework, which efficiently mines high-quality dense subgraphs in dynamic multilayer networks through an incremental update mechanism, achieving significant speed improvements over state-of-the-art methods while maintaining near-optimal density quality.
Using the OULAD dataset, this study demonstrates that while BiLSTM models outperform gradient boosting baselines in predicting student pass/fail and withdrawal outcomes, multi-task learning architectures provide no significant advantage over single-task approaches despite moderate task correlation, suggesting that these outcomes rely on partially distinct behavioral patterns.
This paper presents a systematic review and quantitative meta-analysis of 36 studies demonstrating that tree-based ensemble methods are the most effective supervised machine learning approach for predicting perovskite solar cell efficiency, while also identifying key challenges in data standardization and proposing a comprehensive roadmap to enhance the reliability and scalability of ML-driven photovoltaic development.
This paper introduces ThinkCode, an AI-mediated scaffolding framework that utilizes a Hint-First interaction paradigm to counteract cognitive offloading and skill atrophy in programming education by enforcing active recall and stepwise reasoning before revealing full solutions.
This paper proposes an improved, lightweight YOLOv8-based detection system enhanced with multi-module optimizations to overcome existing limitations in accuracy, real-time performance, and edge-device compatibility for detecting electric bicycle rider helmets in complex traffic scenarios.
The paper introduces the Quantum Kernel Transformer (QKT), a hybrid architecture that leverages a trainable parameterized quantum circuit as a non-linear embedding layer to achieve competitive or superior performance on diverse high-dimensional classification tasks across text, tabular, audio, and physics domains, while demonstrating robustness against instability and identifying specific limitations in image processing.