754 papers
This collection explores the intersection of energy efficiency and evolutionary systems, a rapidly evolving field where researchers investigate how biological principles can inspire more sustainable technologies. By studying natural processes, scientists aim to design algorithms and hardware that consume less power while maintaining high performance, bridging the gap between biology and engineering.
Every new preprint in this category arrives directly from arXiv, and our team at Gist.Science processes each one immediately. We provide both detailed technical summaries for experts and plain-language explanations for anyone curious about the latest breakthroughs, ensuring these complex studies are accessible to a wider audience. Below are the most recent papers in Eess — Iv, ready for you to explore.
This paper proposes an efficient nonlinear contrast source inversion scheme for sparse 2D electromagnetic imaging that directly addresses nonlinearity via nonlinear Landweber iterations enhanced by a self-adaptive projected accelerated steepest descent algorithm to enforce sparsity constraints while ensuring convergence.
This paper demonstrates that frozen DINOv3 vision foundation models, when combined with a face-part labeling interface, enable strong zero-shot region-level facial correspondence and temporal tracking across identities without specialized training, with intermediate feature layers proving most effective for dense anatomical alignment.
This paper proposes a GPU-friendly compression method for 3D Gaussian Splatting that leverages hardware-accelerated texture coding and primitive reordering to efficiently compress spherical harmonic color coefficients with negligible visual quality loss while enabling large-scale parallel decoding.
This paper introduces PTACL, a multimodal contrastive learning framework that leverages both global patient-level and local temporal-level alignment to integrate structural insights from Cardiac MRI into ECG representations, thereby significantly improving cardiac phenotype retrieval and functional parameter prediction without adding new learnable parameters.
This paper proposes a DPU-aware architecture that adapts modern attention-based YOLO variants (YOLOv26 and YOLOv11) for deployment on AMD FPGAs by approximating attention mechanisms and modifying operations for hardware compatibility, achieving up to 3x lower power consumption and high throughput with only a 5% accuracy trade-off across six benchmark datasets.
This paper introduces "Bring Music The Horizon," an emotion-aware pipeline that generates immersive 360° music visualization videos by predicting a song's emotional trajectory and utilizing it to guide keyframe generation and video synthesis for a temporally continuous, immersive experience.
This paper proposes an automated microscopic urinalysis pipeline that reconstructs all-in-focus images from short, manually focused videos to enable efficient deep learning-based detection and classification of urine sediments, thereby eliminating the time-consuming process of capturing multiple discrete focal-plane images.
This study evaluates multi-modal chest radiograph classification using patient-clustered bootstrapping to demonstrate that while prospective clinical indications significantly improve performance, post-hoc report text creates substantial label leakage, and permutation-aware fusion strategies like DeepSets and random-swap offer competitive, well-calibrated alternatives to standard fusion methods.
TCAM-Diff is a novel 3D medical image generation model that reduces memory requirements by combining a decoder-only autoencoder for triplane representation with a triplane-aware cross-attention diffusion model, achieving superior reconstruction and generation quality across various medical datasets compared to existing methods.
This review paper systematically categorizes and evaluates recent Transformer-based and Large Language Model advancements in UAV systems, offering a unified taxonomy, comparative performance analyses, and a critical assessment of challenges and future directions to guide researchers and practitioners in the field.