Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

Methods of Medium Optimization01:28

Methods of Medium Optimization

70
Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...
70

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

A nanozyme-engineered implant surface for sequential photodynamic antibacterial and cytoprotective functions.

Colloids and surfaces. B, Biointerfaces·2026
Same author

Navigating diagnosis: UK informal caregivers' experiences of the dementia assessment journey.

Aging & mental health·2026
Same author

Unlocking an Efficient SOD-to-CAT Catalytic Relay in a MOF Nanozyme via Dopamine-Driven Interfacial Nanoreconstruction for Osteoarthritis Therapy.

Advanced materials (Deerfield Beach, Fla.)·2026
Same author

Injectable curdlan-derived hydrogel with differential release of copper and zinc ions for synergistic therapy of infected wounds.

Carbohydrate polymers·2026
Same author

A 3D-Printed Composite Hydrogel Engineered with a Coordination-Reprogrammed Nanozyme for Synergistic Chemo-Physical Bone Therapy.

ACS applied materials & interfaces·2026
Same author

Unveiling causal relationships between inflammatory cytokines and neurite orientation dispersion and density imaging phenotypes: Insights from bidirectional Mendelian randomization and bibliometric analysis.

Psychiatry and clinical neurosciences·2026

相关实验视频

Updated: May 3, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

7.5K

参数效率技术与完整微调技术的比较:关于多语言新闻文章分类的案例研究.

Olesya Razuvayevskaya1, Ben Wu1, João A Leite1

  • 1Department of Computer Science, The University of Sheffield, Sheffield, United Kingdom.

PloS one
|May 3, 2024
PubMed
概括
此摘要是机器生成的。

像适配器和低级适配 (LoRA) 这样的参数高效的微调技术提供了高效的语言模型培训. 这项研究表明它们在多语言文本分类中的有效性,即使数据有限.

更多相关视频

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

524
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

548

相关实验视频

Last Updated: May 3, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

7.5K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

524
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

548

科学领域:

  • 自然语言处理自然语言处理.
  • 机器学习 机器学习
  • 计算语言学 计算语言学

背景情况:

  • 参数效率微调 (PEFT) 方法,包括适配器和低级适应 (LoRA),提高语言模型培训效率.
  • 之前的研究表明,PEFT可以提高特定分类任务的性能.
  • 本研究扩展了现有工作,通过评估PEFT对分类性能的影响和与完整微调相比的计算成本.

研究的目的:

  • 调查适配器和LoRA对多语言文本分类性能和计算成本的影响.
  • 分析PEFT在各种多语言文本分类任务中的有效性,包括类型,框架和说服检测.
  • 评估PEFT在不同培训场景 (原始多语言数据,英语翻译,仅英语子集) 和跨语言的表现,特别是对于培训数据有限的任务.

主要方法:

  • 对参数效率微调 (PEFT) 技术 (适配器,LoRA) 与完整微调的比较分析.
  • 对具有不同输入长度,类号和难度等级的多语言文本分类任务的评估.
  • 在不同培训数据配置和语言中对PEFT性能进行深入分析.

主要成果:

  • 具有参数效率的微调技术在多语言文本分类中展示了竞争力的性能和效率.
  • 结果强调了PEFT在多标签分类和非并行多语言任务中的适用性.
  • 根据培训数据场景和特定语言,效率有所不同,提供了细微的见解.

结论:

  • 适配器和LoRA是许多多语言文本分类场景的完整微调的可行和高效替代方案.
  • 这些技术对于涉及不同输入长度和有限数据的任务特别有用.
  • 该研究提供了关于在多种多语言NLP研究和应用中应用PEFT的实际指导.