PulseAugur
中
实时 20:03:28
English(EN) ZO-Act: Efficient Zeroth-Order Fine-Tuning via One-Shot Activation-Informed Low-Rank Subspaces

新的ZO-Act方法实现了高效的LLM微调

研究人员开发了ZO-Act,一种用于大型语言模型高效零阶(ZO)微调的新颖方法。该技术将扰动限制在由输入激活感知的低秩子空间内,与现有的ZO方法相比,降低了方差并提高了性能。ZO-Act优化了轻量级系数矩阵,使其适用于量化LLM并与Adam等优化器兼容。实验表明,在Llama-3-8B和OPT-13B等模型上,在各种语言任务中均取得了持续的提升。 AI

影响 该方法可以在内存受限或无法进行反向传播的情况下,实现更高效的大型语言模型微调。

排序理由 该集群描述了一篇详细介绍LLM微调新颖方法的最新研究论文。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新的ZO-Act方法实现了高效的LLM微调

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群描述了一篇详细介绍LLM微调新颖方法的最新研究论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release
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
144 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.LG TIER_1 English(EN) · Xun Dong, Yibo Xu, Naigang Wang, Xin Li, Penghang Yin, Zi Yang ·

    ZO-Act:通过单次激活信息低秩子空间实现高效的零阶微调

    arXiv:2607.01125v1 Announce Type: new Abstract: Zeroth-order (ZO) optimization enables fine-tuning large language models when backpropagation is unavailable or memory-prohibitive, but existing methods often perturb full model weights or randomly constructed low-dimensional subspa…

  2. arXiv cs.LG TIER_1 English(EN) · Zi Yang ·

    ZO-Act:通过单次激活感知低秩子空间实现高效的零阶微调

    Zeroth-order (ZO) optimization enables fine-tuning large language models when backpropagation is unavailable or memory-prohibitive, but existing methods often perturb full model weights or randomly constructed low-dimensional subspaces, yielding high-variance estimates and limite…