PulseAugur
中
实时 06:56:37
English(EN) Direction or Magnitude? Learning Scalars over Task-Calibrated Directions

SCALAR微调方法大幅削减可训练参数,提升效率

研究人员推出了一种新颖的参数高效微调方法SCALAR,与LoRA等现有技术相比,它显著减少了可训练参数数量和内存需求。SCALAR通过仅学习特定于层的标量系数来实现可比的性能,这些系数应用于从任务激活中派生的固定输入基。在各种推理和生成任务上的实验表明了SCALAR的有效性,一项测试显示,仅使用LoRA可训练参数的4%,即可达到与LoRA相似的准确性。对Qwen2.5-7B的系统测量表明,SCALAR提供了20%的峰值训练内存和6%的吞吐量。 AI

影响 提供了一种更有效的大型语言模型微调方法,可能降低定制和部署的门槛。

排序理由 详细介绍参数高效微调新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

SCALAR微调方法大幅削减可训练参数,提升效率

本文如何被排名

Signal score
37 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
详细介绍参数高效微调新方法的学术论文。[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。

报道来源 [1]

  1. arXiv cs.LG TIER_1 English(EN) · Ziqing Wen, Xiaohui Song, Jiahuan Wang, Ping Luo, Junlin Zeng, Dongsheng Li, Kele Xu, Tao Sun ·

    方向还是幅度?学习任务校准方向上的标量

    arXiv:2611.17784v1 Announce Type: new Abstract: Parameter-efficient fine-tuning (PEFT) aims to preserve adaptation quality while reducing training costs. We propose SCALAR, which achieves performance comparable to LoRA with substantially fewer trainable parameters, lower peak tra…