3020 papers
This paper proposes a novel method for selecting a small, performance-representative validation set from limited medical image data to improve model development accuracy, demonstrating through prostate cancer MRI segmentation tasks that this approach yields performance estimates significantly closer to true independent results than traditional random partitioning.
This paper proposes SDD-YOLO11n, a lightweight object detection model for traffic accident recognition that integrates a ShuffleNetV2 backbone, a DW_ADown module, and a DS_SENetV2 module to significantly reduce computational complexity while improving accuracy and real-time performance compared to the original YOLO11n and other mainstream models.
This paper presents a cooperative reduced Variable Neighbourhood Search (rVNS) metaheuristic to optimize the complex Multi-Trip Capacitated Vehicle Routing and Scheduling Problem with Time Windows and Meal Breaks (MTCVRSPTW-MB) for airport catering operations, demonstrating its superior efficiency and mileage reduction over exact methods on real-life instances at San Francisco International Airport.
The Autonomous Research Assistant is a full-stack AI web application that leverages the Semantic Scholar API, a fine-tuned Linear SVM classifier, and Google Gemini 2.5 Flash to automatically discover, classify, summarize, and cite academic papers, thereby reducing manual literature review time by 70%.
This paper empirically validates the application of the SECI knowledge creation model to AI agent memory by demonstrating that transforming conversational history into structured knowledge significantly outperforms raw verbatim storage in long-term retrieval tasks, primarily by mitigating embedding window truncation that causes raw storage to lose approximately 90% of session content.
The paper proposes an Adaptive Mutation Driven Harris Hawk Optimization (AMDHHO) algorithm that integrates Levy flight and dynamic mutation strategies to effectively solve hybrid flow shop scheduling problems with robotic arms, demonstrating significant reductions in maximum completion time and average waiting time compared to genetic algorithms and original scheduling plans.
This study evaluates the classification fidelity of four large language models in categorizing endodontic examination questions across three educational taxonomies, revealing that while models like Gemini and DeepSeek achieve substantial agreement for Bloom's and SOLO taxonomies, all models struggle with Miller's Pyramid, demonstrating that LLM performance is taxonomy-dependent and distinct from answer accuracy.
This study demonstrates that a ResNet-50 classifier enhanced with Grad-CAM and SHAP explanations can surpass human radiologist sensitivity in detecting breast cancer within dense mammograms, achieving 72.2% sensitivity while providing actionable visual insights to support its deployment as a second-reader tool that flags high-confidence cases and defers ambiguous ones to clinicians.
This study proposes a hybrid forecasting model that integrates Singular Spectrum Analysis (SSA) for data decomposition, a backpropagation-trained multilayer perceptron (MLP-BP) for component modeling, and the Sparrow Search Algorithm (SpSA) for parameter optimization, demonstrating superior accuracy in predicting China's civil aviation passenger traffic compared to conventional methods.
This paper proposes a hybrid K-PSO clustering algorithm that combines K-Means and Particle Swarm Optimization to optimize cluster head selection in environmental monitoring wireless sensor networks, significantly improving energy efficiency, extending network lifetime, and enhancing cluster head stability compared to traditional methods.