flâneur

[2012.09852] SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning

arxiv.org · 797 words · saved by 1 readers

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2012.09852] SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning --> Computer Science > Hardware Architecture arXiv:2012.09852 (cs) [Submitted on 17 Dec 2020 ( v1 ), last revised 18 Jul 2024 (this version, v3)] Title: SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning Authors: Hanrui Wang , Zhekai Zhang , Song Han View a PDF of the paper titled SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning, by Hanrui Wang and Zhekai Zhang and Song Han View PDF HTML (experimental) Abstract: The attention me

saved by

related reading