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

Improving Translational Accuracy02:07

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
Improving Translational Accuracy02:07

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation01:24

One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation

This lesson introduces two critical methods in pharmacokinetics, the Wagner-Nelson and Loo-Riegelman methods, used for estimating the absorption rate constant (ka) for drugs administered via non-intravenous routes. The Wagner-Nelson method relates ka to the plasma concentration derived from the slope of a semilog percent unabsorbed time plot. However, it is limited to drugs with one-compartment kinetics and can be impacted by factors like gastrointestinal motility or enzymatic degradation.
On...
Linearization and Approximation01:26

Linearization and Approximation

Linearization is a mathematical technique used to approximate complex, nonlinear functions with simpler linear models in the vicinity of a chosen reference point. The method is based on the idea that, although a function may be difficult to evaluate exactly, its behavior near a specific input value can often be closely approximated by the tangent line at that point. This approach is particularly useful when small deviations from a known value are involved.Consider the square root function, for...
Per-Unit Sequence Models01:26

Per-Unit Sequence Models

An ideal Y-Y transformer, grounded through neutral impedances, displays per-unit sequence networks akin to those of a single-phase ideal transformer when subjected to balanced positive- or negative-sequence currents. These currents do not produce neutral currents, and their associated voltage drops.
Zero-sequence currents, which are identical in magnitude and phase, generate a neutral current, resulting in voltage drops across the neutral impedance and the low-voltage winding. If the...
Methods of Medium Optimization01:28

Methods of Medium Optimization

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...

You might also read

Related Articles

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

Sort by
Same author

YOLOv11-Lite architecture for wildlife detection from drone images.

Frontiers in artificial intelligence·2026
Same author

A quantum resilient deepfake detection framework using enhanced resnext and post quantum cryptography defence.

Scientific reports·2026
Same author

Graph-enhanced multimodal fusion of vascular biomarkers and deep features for diabetic retinopathy detection.

Frontiers in artificial intelligence·2026
Same author

Cytokinin-mediated repression of anthocyanin biosynthesis in banana fruits.

The Plant journal : for cell and molecular biology·2025
Same author

Federated learning with integrated attention multiscale model for brain tumor segmentation.

Scientific reports·2025
Same author

Heat-responsive MaHSF11 transcriptional activator positively regulates flavonol biosynthesis and flavonoid B-ring hydroxylation in banana (Musa acuminata).

The Plant journal : for cell and molecular biology·2025

Related Experiment Video

Updated: Jun 16, 2026

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

U-SplitDoRA: an improved privacy-preserved U-shaped split parameter-efficient fine-tuning framework through weight

Samar Singh1, Brindha Subburaj1, R Alagewaran2

  • 1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India.

Frontiers in Artificial Intelligence
|June 15, 2026
PubMed
Summary

U-SplitDoRA enhances federated learning for large language models (LLMs) by combining split learning and weight decomposition. This privacy-preserving method improves adaptation quality and training efficiency for LLMs in distributed settings.

Keywords:
distributed learninglarge language modelsparameter-efficient fine-tuningsplit federated learningweight-decomposed low-rank adaptation

Related Experiment Videos

Last Updated: Jun 16, 2026

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

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Natural Language Processing

Background:

  • Large language models (LLMs) require efficient fine-tuning methods like parameter-efficient fine-tuning (PEFT) for resource-constrained and privacy-sensitive applications.
  • Federated learning (FL) enables collaborative LLM training without data sharing, addressing data scarcity but facing client-side training overhead.
  • Existing FL LLM fine-tuning methods like SplitLoRA show promise but require improvements in privacy preservation and adaptation quality.

Purpose of the Study:

  • To propose U-SplitDoRA, an improved privacy-preserved U-shaped split parameter-efficient fine-tuning framework for LLMs.
  • To enhance adaptation quality and privacy in federated learning for LLMs using split learning and weight decomposition.
  • To address the training overhead and privacy concerns in collaborative LLM fine-tuning.

Main Methods:

  • U-SplitDoRA utilizes a U-shaped model partitioning strategy (head, body, tail) within a federated learning framework.
  • It employs weight-decomposed low-rank adaptation (DoRA) as the PEFT technique, updating both weight magnitude and direction.
  • The head and tail of the model remain on the client side, while the body is processed on the server, ensuring strong privacy.

Main Results:

  • U-SplitDoRA demonstrated superior accuracy scores and faster convergence speeds compared to state-of-the-art (SOTA) LLM fine-tuning frameworks.
  • The framework effectively reduces the gap between PEFT fine-tuning and full parameter fine-tuning.
  • Experimental results using GPT-2-S and GPT-2-M on the E2E benchmark validated the proposed method's effectiveness.

Conclusions:

  • U-SplitDoRA successfully addresses key challenges in privacy preservation and adaptation quality for federated LLM fine-tuning.
  • The proposed framework offers an efficient, robust, and privacy-preserving solution for fine-tuning LLMs in distributed environments.
  • This research paves the way for more secure and effective collaborative LLM development.