2787 papers
This paper proposes an AI-driven framework for Vehicular Edge Computing that integrates Deep Reinforcement Learning for priority-aware task offloading and LSTM-based resource prediction to dynamically optimize task execution and reduce response time by 31.2% compared to existing methods.
This study proposes an explainable automated lung opacity detection system using a fine-tuned VGG-19 model integrated with Grad-CAM, which achieves 93% accuracy on a Kaggle dataset to support reliable pneumonia diagnosis in children.
This paper proposes BGCE-GeoVis-GNN, a novel framework that enhances the stability of joint video human-object interaction segmentation and recognition by introducing an auxiliary boundary module and collaborative constraints to mitigate boundary shifts and label discontinuities caused by weak transitions, achieving significant performance improvements on MPHOI-72 and CAD-120 datasets.
This paper introduces Columnar-Embedder, a biologically inspired architecture that utilizes local Hebbian learning on streaming random walks to generate robust, binary sparse distributed graph representations capable of competitive performance in node classification and link prediction without requiring backpropagation or suffering from catastrophic forgetting.
This paper introduces ViTone, a diagnostic benchmark and pre-registered evaluation protocol designed to directly assess tonal faithfulness in Vietnamese text-to-speech systems by combining minimal-pair stimuli, an ASR-calibrated Tone Error Rate, and cue-level analysis of F0 and laryngealisation, addressing the limitations of existing metrics that fail to capture the language's contrastive tonal distinctions.
This paper investigates fine-tuning large language models for automatic hint generation in reasoning tasks by comparing four prompting strategies across three models, evaluating hint quality with a new metric suite, and demonstrating that metric-guided dataset merging yields consistent improvements in convergence and familiarity while preserving relevance and minimizing answer leakage.
This study demonstrates that utilizing full-probability readout instead of expectation-value measurements in a compact hybrid quantum self-attention model can improve medical image classification performance on specific datasets, highlighting readout design as a critical architectural factor while noting that the findings do not yet establish quantum advantage or clinical readiness.
This paper presents a deterministic mutation-testing protocol that reveals state-scoped oracles in software-agent benchmarks effectively reject harmful state faults and benign schema variations, while identifying specific false negatives in ToolSandbox where evaluators fail to detect unintended persistent side effects despite unchanged scores.
This paper introduces ADAM, a multi-agent large language model framework that balances high autonomy with targeted human oversight by using semantic reversibility boundaries to safely guide closed-loop materials experimentation.
This survey presents a comprehensive multi-dimensional taxonomy for detecting misbehavior in Cooperative Intelligent Transportation Systems, mapping advanced methodologies like federated learning and blockchain to real-world challenges while addressing the emerging threats of Generative AI and outlining future directions for resilient, privacy-preserving defense mechanisms.