SATIN: a sparse attention application-specific instruction-set processor with custom mixed-precision extensions for energy-efficient edge inference of transformer large language models
🔗 https://www.semanticscholar.org/paper/0742b8811c78b0e55c0ee3496f1c79c965271d36 ↗SATIN: a sparse attention application-specific instruction-set processor with custom mixed-precision extensions for energy-efficient edge inference of transformer large language models. Journal of King Saud University: Computer and Information Sciences
內容為收集當日抓取的摘要原文。