English(EN) Beyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training

新系统自动化大模型训练配置，提升性能

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-14 05:00

研究人员开发了Primus Projection，一个旨在优化分布式大语言模型（LLM）训练复杂配置的新系统。该系统结合了混合投影引擎（测量和模拟GPU性能）和自动化调优代理。调优代理利用投影引擎作为快速评分预言机，探索巨大的配置空间，显著减少了此类过程中通常涉及的试错。Primus Projection在Llama 3.1和Mixtral 8x22B等模型上进行了验证，在性能预测方面表现出高准确性，并在一个案例研究中实现了Mixtral 8x22B的吞吐量提高27%。 AI

影响自动化并加速大模型训练配置，可能降低成本并加快模型开发速度。

排序理由该集群包含一篇详细介绍优化大模型训练新方法的学术论文。[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: 17 / 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优化大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: 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.AI TIER_1 English(EN) · Anshu Raina, Peyman Razaghi, Yuankai Chen, Cheng Yao, Yao Fu, Vidushi Goyal, Devang Patel, Wen Xie, Zhenyu Gu, Emad Barsoum · 2026-11-14 05:00

超越试错循环：混合投影与自动化调优助力分布式训练

arXiv:2610.19423v2 Announce Type: replace Abstract: Configuring distributed LLM training across hundreds of GPUs requires jointly optimizing tensor, pipeline, expert, context, and data parallelism alongside micro-batch size, recomputation, sharding, and overlap, a combinatorial s…

本文如何被排名

报道来源 [1]

超越试错循环：混合投影与自动化调优助力分布式训练

相关实体

相关话题