English(EN) LCT: Losslessly Compressed Tensors with GPU friendly execution for Training

新的LCT框架通过无损压缩大幅降低AI训练内存使用量

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-16 04:00

研究人员开发了无损压缩张量（LCT），一个新颖的GPU执行框架，旨在减少AI模型训练期间的内存使用。LCT利用霍夫曼编码压缩张量，实现了高达1.455倍的压缩率。该方法已在NanoGPT等模型中显示出高达24.6%的内存使用量减少，并能够实现Qwen3-4B等模型微调期间更长的上下文长度。 AI

影响降低了训练和微调的内存需求，有可能在现有硬件上实现更大的模型和上下文长度。

排序理由详细介绍AI模型训练新技术的学术论文。[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: 17 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍AI模型训练新技术的学术论文。[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: 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) · Max Zhu, Bohang Sun, Pietro Lio, Jialin Yu · 2026-10-16 04:00

LCT：用于训练的无损压缩张量，支持GPU友好执行

arXiv:2610.22179v1 Announce Type: new Abstract: Lossless tensor compression could reduce memory usage during training, but training requires tensors to be repeatedly re-encoded. We introduce Losslessly Compressed Tensors (LCT), a GPU execution framework designed for this repeated…