English(EN) MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenarios

MoE-SpAc框架提升边缘设备上MoE模型推理效率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-26 04:00

研究人员开发了MoE-SpAc，一个旨在提高异构边缘设备上混合专家（MoE）模型推理效率的新型框架。该系统通过预测专家激活需求来管理内存，重新利用了推测解码技术。MoE-SpAc集成了推测效用估计器、异构工作负载均衡器和异步执行引擎，以优化预取和驱逐过程。实验表明，与现有的推测解码基线相比，每秒事务处理量提高了42%，并且与其他标准方法相比速度显著提升。 AI

影响提高了在资源受限的边缘设备上运行大型MoE模型的效率。

排序理由该集群包含一篇详细介绍新型高效模型推理框架的研究论文。[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: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍新型高效模型推理框架的研究论文。[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: 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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Shuhuai Li, Jianghao Lin, Dongdong Ge, Yinyu Ye · 2026-10-26 04:00

MoE-SpAc：异构边缘场景下基于推测激活效用的高效MoE推理

arXiv:2603.09983v2 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) models enable scalable performance but face severe memory constraints on edge devices. Existing offloading strategies struggle with I/O bottlenecks due to the dynamic, low-information nature of aut…

本文如何被排名

报道来源 [1]

MoE-SpAc：异构边缘场景下基于推测激活效用的高效MoE推理

相关实体

相关话题