English(EN) MX-SAFE: Versatile Inference- and Training-Proof Microscaling Format with On-the-Fly Exponent and Mantissa Bit Allocation

新的MX-SAFE格式通过自适应量化大幅降低AI能耗

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-26 04:00

研究人员推出了一种新颖的动态量化格式MX-SAFE，旨在降低深度学习的计算成本。该格式通过自适应分配指数和尾数的比特，增强了现有的微缩放（MX）标准，支持训练和推理，并提高了精度。所提出的MX-SAFE格式在现有MXFP格式上平均精度提高了高达3.55%，并在专用加速器上实现了与BF16基线相当的精度，同时能耗降低了24.9%。 AI

影响这种新的量化格式有望显著降低训练和运行AI模型的能耗和计算成本。

排序理由该集群包含一篇学术论文，详细介绍了提高AI硬件效率的新技术格式。[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; 该集群包含一篇学术论文，详细介绍了提高AI硬件效率的新技术格式。[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: paper, infra; 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: 97 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) · Dahoon Park, Jahyun Koo, Sangwoo Hwang, Jaeha Kung · 2026-05-26 04:00

MX-SAFE：一种多功能、可防止推理和训练的微缩格式，具有动态指数和尾数比特分配

arXiv:2605.24391v1 Announce Type: cross Abstract: As the demand for deep learning grows, cost reduction through quantization has become essential for both training and inference. In 2022, the Open Compute Project (OCP) consortium standardized narrow precision formats for deep lea…

本文如何被排名

报道来源 [1]

MX-SAFE：一种多功能、可防止推理和训练的微缩格式，具有动态指数和尾数比特分配

相关实体

相关话题