English(EN) Cleave: Scaling Tensor Program Optimization via Decoupled Algebraic Search and Operator Scheduling

Cleave编译器优化大型模型的张量程序

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-07 04:00

研究人员开发了Cleave，这是一种新的机器学习编译器，旨在优化大型模型的张量程序。Cleave采用符号解耦策略，首先在具有符号形状的图上执行超优化，然后在具体形状上调度转换后的图。这种方法可以有效地处理具有多个归约的计算，并已显示出显著的速度提升，生成的内核比现有基线快2.8倍。 AI

影响 Cleave的优化技术可能导致大型AI模型执行更快、更高效。

排序理由详细介绍用于优化ML模型的新编译器技术的学术论文。[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: 19 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍用于优化ML模型的新编译器技术的学术论文。[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: 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.AI TIER_1 English(EN) · David Pissarra, Jinkun Lin, Haitian Jiang, Aurojit Panda, Jinyang Li · 2026-10-07 04:00

Cleave：通过解耦代数搜索和算子调度实现张量程序优化的扩展

arXiv:2610.07742v1 Announce Type: cross Abstract: Optimized kernels such as FlashAttention and FlashDecoding are crucial for accelerating today's large models. Most of them are handwritten by experts because existing ML compilers cannot match their efficiency. Producing such kern…