English(EN) HAPMoE: Heterogeneity-Aware Automatic Parallelism Planning for Mixture-of-Experts Models Training

新的HAPMoE系统优化了异构集群上的MoE模型训练

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-02 04:00

研究人员开发了HAPMoE，一个旨在优化大型专家混合（MoE）模型在异构计算集群上训练的新系统。该方法解决了在不同硬件上高效并行化MoE架构的挑战，而现有方法未能充分解决此问题。HAPMoE利用一个MoE感知型成本模型和动态规划算法来搜索最优并行策略，显著提高了训练吞吐量。 AI

影响优化大型MoE模型的训练，可能降低计算成本并加速开发。

排序理由该集群包含一篇详细介绍AI模型训练新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

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: 13 / 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AI模型训练新方法的学术论文。[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: 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。

arXiv cs.LG TIER_1 English(EN) · Mengyuan Fan, Peizhuang Cong, Zixiao Huang, Si Xu, Tong Qiao, Yanghao Li, Jing Yang, Tong Yang, Quanlu Zhang, Yu Wang · 2026-10-02 04:00

HAPMoE：面向混合专家模型训练的异构感知自动并行规划

arXiv:2609.39350v1 Announce Type: cross Abstract: As model sizes continue to scale, distributed training has become inevitable. Automatic parallelization techniques can derive efficient training parallelism strategies at low cost while achieving superior performance. The difficul…