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

Protein Networks02:26

Protein Networks

2.4K
2.4K

You might also read

Related Articles

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

Sort by
Same author

Lipid-rich necrotic core of the carotid plaque and the risk of major adverse cardiovascular and cerebrovascular events: a meta-analysis and systematic review.

PeerJĀ·2026
Same author

Key Toxic Genes and Mechanisms of Di(2-ethylhexyl) Phthalate-Induced Non-Alcoholic Fatty Liver Disease: A Multi-Omics Study.

Current medicinal chemistryĀ·2026
Same author

Efficient Tuning Framework for Resource- Constrained Biomedical Question Answering.

IEEE transactions on computational biology and bioinformaticsĀ·2026
Same author

Cu-Co Dual Single Atom and Nitrogen Doped Carbon Nanotubes as Oxygen Reduction Reaction Electrocatalysts.

ACS applied materials & interfacesĀ·2026
Same author

Control of the cardiovascular risk factors in patients with type 2 diabetes mellitus in China in 2023-2024.

Diabetes, obesity & metabolismĀ·2025
Same author

Symmetry-Breaking Magneto-Optical Effects in Altermagnets.

Nano lettersĀ·2025

Related Experiment Video

Updated: Sep 28, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.3K

UGDAS: Unsupervised graph-network based denoiser for abstractive summarization in biomedical domain.

Yongping Du1, Yiliang Zhao1, Jingya Yan1

  • 1Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China.

Methods (San Diego, Calif.)
|April 4, 2022
PubMed
Summary

This study introduces UGDAS, a novel model for abstractive summarization that effectively denoises text using graph networks and domain knowledge. The model achieves state-of-the-art results on the COVID-19 Open Research Dataset (CORD-19).

Keywords:
Abstractive summarizationDomain knowledgeGraph-networkPre-trained language model

More Related Videos

Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
03:37

Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers

Published on: March 1, 2024

929
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

629

Related Experiment Videos

Last Updated: Sep 28, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.3K
Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
03:37

Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers

Published on: March 1, 2024

929
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

629

Area of Science:

  • Natural Language Processing
  • Artificial Intelligence
  • Machine Learning

Background:

  • Abstractive summarization models often struggle with text noise, impacting summary quality.
  • Graph-based networks are effective for capturing cross-sentence relationships crucial for summarization.
  • Domain-specific knowledge is vital for identifying and mitigating noise in specialized texts.

Purpose of the Study:

  • To propose a novel model structure, UGDAS, for abstractive summarization.
  • To improve summary generation quality by denoising input text.
  • To leverage domain knowledge and sentence position for enhanced text preprocessing.

Main Methods:

  • Developed UGDAS, combining an unsupervised graph-network based sentence-level denoiser with an auto-regressive generator.
  • Utilized domain knowledge and sentence position information within the denoising process.
  • Employed the COVID-19 Open Research Dataset (CORD-19) and PubMed Abstract dataset for evaluation.

Main Results:

  • Achieved state-of-the-art (SOTA) performance on the CORD-19 dataset.
  • Outperformed existing baseline models on the PubMed Abstract dataset.
  • Demonstrated the effectiveness of the UGDAS model in improving abstractive summarization quality.

Conclusions:

  • The proposed UGDAS model significantly enhances abstractive summarization by effectively denoising text.
  • The integration of unsupervised graph networks and domain knowledge proves beneficial for summarization tasks.
  • UGDAS represents a significant advancement in generating high-quality summaries from noisy, domain-specific corpora.