Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Ranks01:02

Ranks

505
Unlike parametric methods, nonparametric statistics are ideal for nominal and ordinal data, requiring fewer assumptions about the population's nature or distribution. This makes nonparametric methods easier to apply and interpret, as they do not depend on parameters like mean or standard deviation. One common approach in nonparametric analysis is to sort data according to a specific criterion. For instance, we might arrange weather data from hottest to coldest days in a month or rank cities...
505
Spearman's Rank Correlation Test01:20

Spearman's Rank Correlation Test

1.5K
Spearman's rank correlation test, also known as Spearman's rho, is a nonparametric method for assessing the strength and direction of association between two variables. This test is particularly valuable when the data distribution is unknown or when the assumption of normality does not hold. Named after the English psychologist and statistician Dr. Charles Edward Spearman, it serves as the nonparametric counterpart to Pearson's correlation coefficient.
Spearman's test calculates correlation by...
1.5K
Mechanism of Breathing I: Inspiration01:30

Mechanism of Breathing I: Inspiration

3.2K
Introduction to Inspiration: The Respiratory System in Action
The respiratory system, an essential network for breathing, comprises the conducting and respiratory zones, each playing a crucial role in the overall process of respiration. Let us explore the detailed mechanism of inspiration, or inhalation, which is the first phase of the respiratory cycle.
Pathway of Air during Inspiration
During inspiration, air enters our body through the nose or mouth and moves through the conducting zone,...
3.2K
Fineness of Cement01:15

Fineness of Cement

518
The fineness of cement directly influences the rate of hydration, as the hydration begins at the surface of the cement particles. In addition to hydration, the fineness of cement is vital for various properties of concrete including workability, gypsum requirement, and long-term behavior. The fineness of cement is represented in terms of the specific surface of cement which is typically measured in square meters per kilogram, with several methods available for this determination.
Direct...
518
Fineness Modulus01:19

Fineness Modulus

1.5K
The fineness modulus (FM) of aggregate is a numerical index that measures the coarseness or fineness of the particles. It is calculated by adding the cumulative percentages of aggregate retained on each of a specified series of sieves and dividing the sum by 100.
Consider performing sieve analysis on sand through a set of ASTM sieves. The weight of aggregate retained in each sieve and pan placed at the bottom is recorded, as given in Column B of Table 1.
To determine the fineness modulus of...
1.5K
Wilcoxon Rank-Sum Test01:21

Wilcoxon Rank-Sum Test

755
The Wilcoxon rank-sum test, also known as the Mann-Whitney U test, is a nonparametric test used to determine if there is a significant difference between the distributions of two independent samples. This test is designed specifically for two independent populations and has the following key requirements:
755

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Healthcare system cost evaluation of antiviral stockpiling for pandemic influenza preparedness.

Biosecurity and bioterrorism : biodefense strategy, practice, and science·2010
Same author

PDGF-CC blockade inhibits pathological angiogenesis by acting on multiple cellular and molecular targets.

Proceedings of the National Academy of Sciences of the United States of America·2010
Same author

Betulin induces mitochondrial cytochrome c release associated apoptosis in human cancer cells.

Molecular carcinogenesis·2010
Same author

Development of novel 5-fluorouracil carrier erythrocyte with pharmacokinetics and potent antitumor activity in mice bearing malignant ascites.

Journal of gastroenterology and hepatology·2010
Same author

[Effectiveness of educational interventions in children with chronic diseases and their parents].

Zhongguo dang dai er ke za zhi = Chinese journal of contemporary pediatrics·2010
Same author

Genomic identification of a novel mutation in hfq that provides multiple benefits in evolving glucose-limited populations of Escherichia coli.

Journal of bacteriology·2010

Related Experiment Video

Updated: Feb 3, 2026

Dithranol as a Matrix for Matrix Assisted Laser Desorption/Ionization Imaging on a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer
09:38

Dithranol as a Matrix for Matrix Assisted Laser Desorption/Ionization Imaging on a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer

Published on: November 26, 2013

14.6K

Matrix-Transformation based Low-Rank Adaptation (MTLoRA): A brain-inspired method for parameter-efficient

Yao Liang1, Yuwei Wang2, Yang Li1

  • 1Brain-inspired Cognitive AI Lab, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China; School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, 100049, China.

Neural Networks : the Official Journal of the International Neural Network Society
|February 1, 2026
PubMed
Summary

Matrix-Transformation based Low-Rank Adaptation (MTLoRA) enhances parameter-efficient fine-tuning by learning data-adapted geometry within low-rank subspaces. This approach improves large language model performance and training stability over standard methods like LoRA.

Keywords:
Fine-tuningLarge language modelsLarge modelsLow-rank matricesParameter-efficient fine-tuning

More Related Videos

Author Spotlight: Streamlining Composite Plant Production via Agrobacterium rhizogenes-Mediated Hairy Root Transformation
04:09

Author Spotlight: Streamlining Composite Plant Production via Agrobacterium rhizogenes-Mediated Hairy Root Transformation

Published on: June 30, 2023

3.0K
Tuning Degradation to Achieve Specific and Efficient Protein Depletion
05:11

Tuning Degradation to Achieve Specific and Efficient Protein Depletion

Published on: July 20, 2019

6.6K

Related Experiment Videos

Last Updated: Feb 3, 2026

Dithranol as a Matrix for Matrix Assisted Laser Desorption/Ionization Imaging on a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer
09:38

Dithranol as a Matrix for Matrix Assisted Laser Desorption/Ionization Imaging on a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer

Published on: November 26, 2013

14.6K
Author Spotlight: Streamlining Composite Plant Production via Agrobacterium rhizogenes-Mediated Hairy Root Transformation
04:09

Author Spotlight: Streamlining Composite Plant Production via Agrobacterium rhizogenes-Mediated Hairy Root Transformation

Published on: June 30, 2023

3.0K
Tuning Degradation to Achieve Specific and Efficient Protein Depletion
05:11

Tuning Degradation to Achieve Specific and Efficient Protein Depletion

Published on: July 20, 2019

6.6K

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Natural Language Processing

Background:

  • Parameter-efficient fine-tuning (PEFT) methods like LoRA reduce computational costs for adapting large language models (LLMs).
  • Standard low-rank adapters can underperform full fine-tuning due to limitations in updating a fixed rank-r subspace.
  • There is a need for PEFT techniques that enhance performance and stability without significantly increasing computational overhead.

Purpose of the Study:

  • To introduce Matrix-Transformation based Low-Rank Adaptation (MTLoRA), a novel PEFT method.
  • To improve the performance and stability of LLM fine-tuning by enabling data-adapted geometric transformations within the low-rank subspace.
  • To provide a plug-compatible and efficient alternative to existing low-rank adaptation techniques.

Main Methods:

  • MTLoRA inserts a learnable r x r transformation matrix (T) into the low-rank update (ΔW=BTA).
  • Four specific structures for T (SHIM, ICFM, CTCM, DTSM) are proposed, offering distinct inductive biases.
  • Optimization analysis demonstrates T's role as a learned preconditioner, enhancing training stability.

Main Results:

  • MTLoRA consistently improves performance across various tasks and models, including GLUE, natural language generation, and multimodal instruction tuning.
  • On GLUE with DeBERTaV3-base, MTLoRA improved average scores by 2.0 points over LoRA.
  • In multimodal tuning with LLaVA-1.5-7B, a DTSM variant achieved the best average score with minimal trainable parameters, outperforming full fine-tuning.

Conclusions:

  • Learning geometric transformations within the low-rank subspace significantly enhances both the effectiveness and stability of LLM fine-tuning.
  • MTLoRA offers a practical and efficient alternative to standard LoRA, achieving superior results with comparable or lower computational demands.
  • The proposed method advances the field of parameter-efficient adaptation for large-scale AI models.