English(EN) Cache-Aware Joint Router Adaptation for Memory-Efficient MoE Inference

新框架提升MoE模型推理效率

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-07 04:00

研究人员开发了一个感知缓存的框架，以提高混合专家（MoE）模型在推理过程中的内存效率。所提出的训练后方法联合自适应MoE骨干网络和轻量级辅助缓存路由器，旨在减少当完整专家集超出GPU内存时重复权重传输的需求。在Qwen3和GPT-OSS模型上评估了时间路由器和时空路由器两种模式，显示缓存命中率显著提高，专家权重流量减少。 AI

影响这项研究可能导致更有效地部署大型混合专家模型，降低硬件需求和推理成本。

排序理由该集群包含一篇详细介绍改进AI模型推理新技术的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 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: 33 / 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改进AI模型推理新技术的学术论文。[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: 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.CL TIER_1 English(EN) · Zhenhe Wu, Yaping Jin, Qinghua Xing, Hang Zhou, Wei He, Xianjie Wu, Xianfu Cheng, Jian Yang, Hanting Chen · 2026-09-07 04:00

面向内存高效MoE推理的缓存感知联合路由器自适应

arXiv:2609.04895v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models activate only a small subset of experts per token, but the full expert set often exceeds GPU memory, causing repeated weight transfers during decoding. We formulate expert-cache management as a model-…

本文如何被排名

报道来源 [1]

面向内存高效MoE推理的缓存感知联合路由器自适应

相关实体

相关话题