Results 131 to 140 of about 724,897 (308)

The scheduling of sparse matrix-vector multiplication on a massively parallel dap computer [PDF]

, 1991
An efficient data structure is presented which supports general unstructured sparse matrix-vector multiplications on a Distributed Array of Processors (DAP). This approach seeks to reduce the inter-processor data movements and organises the operations in
Mitra, G, Parkinson, D, Andersen, J
core   +2 more sources

Dynamic Sparse Training of Diagonally Sparse Networks

CoRR
Recent advances in Dynamic Sparse Training (DST) have pushed the frontier of sparse neural network training in structured and unstructured contexts, matching dense-model performance while drastically reducing parameter counts to facilitate model scaling. However, unstructured sparsity often fails to translate into practical speedups on modern hardware.
Abhishek Tyagi, Arjun Iyer, William H. Renninger, Christopher Kanan, Yuhao Zhu 0001   +4 more
openaire   +3 more sources

Composites of Shellac and Silver Nanowires as Flexible, Biobased, and Corrosion‐Resistant Transparent Conductive Electrodes

Advanced Functional Materials, EarlyView.
Shellac, a centuries‐old natural resin, is reimagined as a green material for flexible electronics. When combined with silver nanowires, shellac films deliver transparency, conductivity, and stability against humidity. These results position shellac as a sustainable alternative to synthetic polymers for transparent conductors in next‐generation ...
Rahaf Nafez Hussein, Tiago Carneiro Gomes, Eliza Ng, Simon Rondeau‐Gagné, Tricia Breen Carmichael   +4 more
wiley   +1 more source

Electrospun Wood‐Derived Biopolymers as Electrodes in Electrochemical Energy Storage Technologies

Advanced Functional Materials, EarlyView.
Electrospinning transforms wood‐derived cellulose and lignin into architecturally defined, binder‐free carbon electrodes with tuneable porosity and functionality. This review shows how fibre design enables decoupled charge and mass transport, enhancing performance across battery systems, while identifying key challenges in spinnability, scalability ...
Michael W. Thielke, Hattie R. Chisnall, Jesus Vazquez Chavez, Ana B. Jorge Sobrido   +3 more
wiley   +1 more source

Strong Lead‐Free Bioinspired Piezoceramics for Durable Energy Transducers

Advanced Functional Materials, EarlyView.
We fabricated brick‐and‐mortar lead‐free piezoceramics with Bi0.5Na0.5TiO3 monocrystal platelets as bricks and amorphous silica as mortar. The alignment of the bricks and the presence of the amorphous silica led to anisotropic residual stresses. These residual stresses increase the samples toughness, strength and durability while the brick alignment ...
Ruxue Yang, Temesgen Tadeyos Zate, Peiren Wang, Soumyajit Mojumder, Elo Overgaard Mogensen, Oriol Gavalda‐Diaz, Zihe Li, Shitong Zhou, Ajeet Kumar, James Roscow, Hamideh Khanbareh, Astri Bjørnetun Haugen, Florian Bouville   +12 more
wiley   +1 more source

SparseD: Sparse Attention for Diffusion Language Models

CoRR
While diffusion language models (DLMs) offer a promising alternative to autoregressive models (ARs), existing open-source DLMs suffer from high inference latency. This bottleneck is mainly due to the attention's quadratic complexity with respect to context length in computing all query-key pairs.
Zeqing Wang, Gongfan Fang, Xinyin Ma, Xingyi Yang, Xinchao Wang   +4 more
openaire   +2 more sources

Dynamically Actuated Reconfigurable Topographical Surface Enables Active Control of Implant‐Associated Infections

Advanced Functional Materials, EarlyView.
A dynamically actuated reconfigurable topographical surface (DARTS) integrates contact‐mediated bactericidal nanotopography with programmable mechanical actuation to actively disrupt bacterial biofilms. Dynamic surface reconfiguration enhances bacterial killing and suppresses implant‐associated infections in vivo, providing a new strategy for active ...
Mohammad Asadi Tokmedash, Jihun Lee, J. Scott VanEpps, Sungmin Nam, Jouha Min   +4 more
wiley   +1 more source

Sparse-ProxSkip: Accelerated Sparse-to-Sparse Training in Federated Learning

In Federated Learning (FL), both client resource constraints and communication costs pose major problems for training large models. In the centralized setting, sparse training addresses resource constraints, while in the distributed setting, local training addresses communication costs.
Meinhardt, Georg, Yi, Kai, Condat, Laurent, Richtárik, Peter   +3 more
openaire   +2 more sources

Gradient‐Conductivity Architecture Enables Synchronous Enhancement of Sensitivity and Detection Range in Multifunctional All‐Paper Sensors

Advanced Functional Materials, EarlyView.
A gradient‐engineered all‐paper sensor is fabricated by integrating MXene and in situ grown AgNPs within hierarchical cellulose networks. The device breaks the sensitivity–detection range trade‐off through cascaded conductive pathways, enabling ultrahigh pressure sensitivity, humidity–pressure decoupled dual‐mode sensing, and outstanding EMI shielding,
Ao Li, Jun Xu, Dezhong Xu, Zhaohui Zhang, Jiyi Li, Daxian Cao, Kefu Chen, Mizi Fan   +7 more
wiley   +1 more source

A Continuous Low-Rank Tensor Approach for Removing Clouds from Optical Remote Sensing Images

Remote Sensing
Optical remote sensing images are often partially obscured by clouds due to the inability of visible light to penetrate cloud cover, which significantly limits their subsequent applications. Most existing cloud removal methods formulate the problem using
Dong-Lin Sun, Teng-Yu Ji, Siying Li, Zirui Song   +3 more
doaj   +1 more source