English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The increasing success and scaling of Deep Learning models demands highercomputational efficiency and power. Sparsification can lead to both smallermodels as well as higher compute efficiency, and accelerated hardware isbecoming available. However, exploiting it efficiently requires kernelimplementations, pruning algorithms, and storage formats, to utilize hardwaresupport of specialized sparse vector units. An example of those are theNVIDIA's Sparse Tensor Cores (SPTCs), which promise a 2x speedup. However,SPTCs only support the 2:4 format, limiting achievable sparsity ratios to 50%.We present the V:N:M format, which enables the execution of arbitrary N:Mratios on SPTCs. To efficiently exploit the resulting format, we proposeSpatha, a high-performance sparse-library for DL routines. We show that Spathaachieves up to 37x speedup over cuBLAS. We also demonstrate a second-orderpruning technique that enables sparsification to high sparsity ratios withV:N:M and little to no loss in accuracy in modern transformers.

VENOM: A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores