English(EN) MIVE: A Minimalist Integer Vector Engine for Softmax LayerNorm and RMSNorm Acceleration

新型 MIVE 引擎加速大语言模型归一化操作

作者 PulseAugur 编辑部 · [1 个来源] · 2026-06-17 04:00

研究人员开发了一种名为 MIVE（Minimalist Integer Vector Engine，极简整数向量引擎）的新硬件架构，旨在加速大语言模型（LLMs）中的关键操作。MIVE 是一种可编程引擎，可以在单个数据通路中高效处理 Softmax、LayerNorm 和 RMSNorm 功能，从而减少对重复硬件资源的需求。ASIC 实现表明，与现有针对这些操作的独立加速器相比，MIVE 在面积和硬件效率方面均有所提高。 AI

影响 MIVE 的高效硬件设计有望实现更快、更节能的大语言模型推理。

排序理由该集群描述了一篇详细介绍用于加速大语言模型操作的新型硬件架构的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一篇详细介绍用于加速大语言模型操作的新型硬件架构的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: infra, paper; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 113 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Kosmas Alexandridis, Giorgos Dimitrakopoulos · 2026-06-17 04:00

MIVE：用于 Softmax LayerNorm 和 RMSNorm 加速的极简整数向量引擎

arXiv:2606.17781v1 Announce Type: cross Abstract: The rapid growth of Large Language Models (LLMs) has intensified the need for specialized hardware accelerators that can satisfy stringent inference latency and power constraints. Although matrix multiplications dominate the overa…

本文如何被排名

报道来源 [1]

MIVE：用于 Softmax LayerNorm 和 RMSNorm 加速的极简整数向量引擎

相关实体

相关话题