2982 papers
This paper proposes an adaptive framework for volumetric density field sequences that utilizes a novel structural descriptor for optimal keyframe selection via dynamic programming and a particle-based interpolation scheme to achieve high-fidelity reconstruction with minimal artifacts.
This paper proposes a lightweight, real-time detector for unmanned surface vehicles that enhances weak water-surface target detection by integrating frequency-guided feature routing and scale-normalized adaptive localization supervision to address challenges like low contrast and small object size while maintaining strict latency and model-size constraints.
This paper introduces a closed-loop Auto Research framework that uses language-model agents to edit molecular representations, model code, and external evidence, successfully discovering and certifying generalizable improvements across multiple molecular property benchmarks while distinguishing true transferable gains from non-transferable selection variance and distribution shifts through rigorous held-out testing.
This paper proposes an explainable AI framework that combines transfer learning, Vision Transformers, and ensemble methods to achieve superior breast cancer prediction accuracy (96.82%) on the BreakHis dataset while utilizing techniques like Grad-CAM and LIME to enhance model transparency and interpretability.
The paper introduces R-Super, a novel framework that leverages over 100,000 radiology reports to train tumor segmentation models without requiring scarce manual masks, thereby enabling large-scale multi-cancer detection that significantly outperforms both existing AI models and human radiologists.
This paper introduces SonicPlay, an Unreal Engine-based framework that operationalizes audio-based accessibility principles into reusable tools, demonstrating through dual studies with blind players and professional developers that it effectively enables independent spatial navigation while seamlessly integrating into standard game development workflows.
This empirical study analyzes data from 145 UCI students using 18 machine learning algorithms to identify that the LightGBM model most accurately predicts academic performance, revealing that personal and school education factors (such as gender, age, part-time work, and class listening) are the primary drivers of success while family factors have a weaker impact, thereby enabling the development of an interpretable Shiny application for personalized academic interventions.
This paper introduces the Collective Adaptation Testbed (CAT), a multi-agent software architecture grounded in Dual-Inheritance Theory that resolves the governance paradox of quantitative metrics by intercepting team decisions to distinguish between blind conformity and expertise-driven dissent, thereby enabling the system to autonomously revise its own rules based on retrospective outcome data.
This paper proposes an artifact-aware lightweight U-shaped segmentation framework and a reproducible corruption benchmark to enhance the robustness and accuracy of dermoscopic lesion segmentation in the presence of acquisition artifacts like hair and specular highlights.
This paper proposes a privacy-preserving deepfake detection framework that integrates Federated Learning with ResNet-18 and the FedProx algorithm to achieve high accuracy (97.20%) on the FaceForensics++ dataset while ensuring data security and model interpretability without transferring raw images.