2786 papers
This paper proposes a "Sheaf-State Language Model" architecture that replaces computationally excessive dense self-attention with a lower-complexity framework using local state-space dynamics and sparse, typed gluing morphisms to efficiently manage context and dependencies.
GaussCast is a dependency-aware delivery framework for multi-user layered 3D Gaussian Splatting that aggregates concurrent user demands at an edge proxy to fetch shared prerequisite blocks only once, thereby significantly reducing upstream traffic and improving startup latency and rendering quality compared to independent per-user delivery strategies.
This paper proposes the State-Sensing Adaptive Artificial Bee Colony (SSA-ABC) algorithm, which overcomes standard ABC's limitations through dimensionality-aware initialization, dynamic search adjustment, and rank-based selection mechanisms to achieve superior performance in high-dimensional optimization and robot path planning.
This study demonstrates that visual distances derived from large vision models significantly correlate with expert-coded historical-artistic relatedness across diverse cultural units and epochs, suggesting that internet-scale visual training corpora inherently encode statistical regularities reflecting art-historical relationships.
This paper proposes a Constrained Soft Actor-Critic (CSAC) approach that decouples stringent delay constraints from the reward function and employs a two-stage mapping mechanism to effectively maximize throughput utility while minimizing delay violations in heterogeneous queuing systems with bursty traffic, outperforming both unconstrained learning and heuristic baselines.
This study demonstrates that among YOLO-family models for femur segmentation in T1-weighted MRI, YOLOv11m-seg provides the optimal balance of data efficiency, accuracy, and speed in data-scarce regimes, achieving high performance with as few as 100 annotated slices while showing that architectural differences become negligible with larger datasets.
This study proposes a computationally efficient 2D image compression method using the Haar Discrete Wavelet Transform to maximize compression ratios while preserving the clinical relevance and diagnostic reliability of physiological images.
BioSurveillance-Agent is a multi-method explainable agentic platform that integrates One Health networks, wearable sensors, and a hybrid AI pipeline to enable context-adaptive, hierarchical epidemic surveillance and intervention, demonstrating significant reductions in disease spread and mortality through robust, human-auditable decision-making.
The paper proposes the Cascaded Multi-Token Disambiguation Model (CMTD), which enhances lip reading performance by integrating future-context modeling directly into the decoding process through a cascaded multi-token prediction structure and a hierarchical training objective, thereby effectively resolving visual ambiguities and reducing error rates on benchmark datasets.
This paper empirically demonstrates that while traditional and non-structural test adequacy metrics show weak or no correlation with mutation scores, Latent Space Class Dispersion (LSCD) serves as a significantly stronger and computationally efficient proxy for evaluating Deep Neural Network test dataset quality compared to Mahalanobis Distance-based Surprise Coverage.