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

Genetic Lingo01:11

Genetic Lingo

104.7K
Overview
104.7K
Multi-species Conserved Sequences02:51

Multi-species Conserved Sequences

4.3K
Next-generation sequencing technologies have created large genomic databases of a variety of animals and plants. Ever since the human genome project was completed, scientists studied the genome of primates, mammals, and other phylogenetically distant living beings. Such large-scale  studies have provided new insights into the evolutionary relationship between organisms.
Although the genome of each species varies greatly from each other, a few sequences are highly conserved. Such conserved...
4.3K
Genomics02:02

Genomics

37.5K
Genomics is the science of genomes: it is the study of all the genetic material of an organism. In humans, the genome consists of information carried in 23 pairs of chromosomes in the nucleus, as well as mitochondrial DNA. In genomics, both coding and non-coding DNA is sequenced and analyzed. Genomics allows a better understanding of all living things, their evolution, and their diversity. It has a myriad of uses: for example, to build phylogenetic trees, to improve productivity and...
37.5K

You might also read

Related Articles

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

Sort by
Same author

Long-term cardiovascular risk in survivors of hematologic malignancies: a meta-analysis.

BMC cancer·2026
Same author

Psychometric performance and student perceptions of AI- versus student-generated multiple-choice questions: a single-center randomized controlled trial.

BMC medical education·2026
Same author

Let large language models judge each other: multi-agent peer-reviewed reasoning for medical question answering.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Development and nested cross-validation of machine learning models for predicting transfusion requirements in pediatric craniosynostosis surgery.

Journal of plastic, reconstructive & aesthetic surgery : JPRAS·2026
Same author

Transcatheter Aortic Valve Replacement in Aortic Regurgitation Secondary to Aortic Valvulitis.

JACC. Case reports·2026
Same author

Divergent Soil Aggregate Stability Despite Similar Organic Carbon Gains Under Long-Term Maize Intercropping with Different Legume Cover Crops.

Microorganisms·2026

Related Experiment Video

Updated: Sep 12, 2025

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

682

MMRAG: multi-mode retrieval-augmented generation with large language models for biomedical in-context learning.

Zaifu Zhan1, Jun Wang2, Shuang Zhou2

  • 1Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, United States.

Journal of the American Medical Informatics Association : JAMIA
|August 5, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces a novel framework to improve example selection for biomedical natural language processing (NLP) tasks. The multi-mode retrieval-augmented generation (MMRAG) framework enhances in-context learning and shows significant performance gains in relation extraction.

Keywords:
in-context learninglarge language modelretrieval augmented generation

More Related Videos

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

579
Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

673

Related Experiment Videos

Last Updated: Sep 12, 2025

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

682
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

579
Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

673

Area of Science:

  • Biomedical Natural Language Processing
  • Machine Learning
  • Artificial Intelligence

Background:

  • In-context learning in biomedical NLP is crucial for tasks like Named Entity Recognition (NER), Relation Extraction (RE), and Text Classification (TC).
  • Effective example selection within prompts significantly impacts the performance of large language models (LLMs) in these specialized domains.
  • Data scarcity and the need for nuanced understanding pose challenges for current biomedical NLP approaches.

Purpose of the Study:

  • To optimize in-context learning for biomedical natural language processing (NLP) by developing an advanced example selection strategy.
  • To introduce and evaluate a novel Multi-Mode Retrieval-Augmented Generation (MMRAG) framework designed to enhance LLM performance on biomedical text analysis tasks.
  • To assess the impact of different retrieval strategies on the efficacy of MMRAG across various biomedical NLP benchmarks.

Main Methods:

  • The study proposes the MMRAG framework, integrating four distinct retrieval strategies: Random, Top, Diversity, and Class modes.
  • MMRAG was evaluated on three core biomedical NLP tasks: NER (BC2GM dataset), RE (DDI and GIT datasets), and TC (HealthAdvice dataset).
  • Experiments utilized Llama-2-7B and Llama-3-8B LLMs with three retrievers (Contriever, MedCPT, BGE-Large) to compare retrieval strategy effectiveness.

Main Results:

  • The Random Mode indicated that increasing prompt examples boosts generation performance.
  • Top Mode and Diversity Mode significantly outperformed Random Mode on the RE (DDI) task, achieving an F1 score of 0.9669 (a 26.4% improvement).
  • Contriever demonstrated superior performance across more experiments compared to MedCPT and BGE-Large; Llama 3 showed an advantage in NER tasks.

Conclusions:

  • The MMRAG framework effectively enhances biomedical in-context learning through optimized example selection.
  • This approach mitigates data scarcity challenges and improves the adaptability of NLP models for healthcare applications.
  • MMRAG demonstrates significant potential for advancing NLP-driven solutions in the biomedical and healthcare sectors.