Jove
Visualize
Contáctanos
JoVE
x logofacebook logolinkedin logoyoutube logo
ACERCA DE JoVE
Visión GeneralLiderazgoBlogCentro de Ayuda JoVE
AUTORES
Proceso de PublicaciónConsejo EditorialAlcance y PolíticasRevisión por ParesPreguntas FrecuentesEnviar
BIBLIOTECARIOS
TestimoniosSuscripcionesAccesoRecursosConsejo Asesor de BibliotecasPreguntas Frecuentes
INVESTIGACIÓN
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchivo
EDUCACIÓN
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualCentro de Recursos para ProfesoresSitio de Profesores
Términos y Condiciones de Uso
Política de Privacidad
Políticas

Videos de Conceptos Relacionados

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems05:47

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

1.3K
This article describes RUGGED (Retrieval Under Graph-Guided Explainable disease Distinction), which integrates Large Language Model (LLM) inference with Retrieval-Augmented Generation (RAG). It draws evidence from expert-curated biomedical knowledge bases and peer-reviewed biomedical publications to synthesize new knowledge from up-to-date information, identify explainable and actionable predictions, and pinpoint promising directions for hypothesis-driven...
1.3K
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports07:35

A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

2.1K
A computational protocol, CaseOLAP LIFT, and a use case are presented for investigating mitochondrial proteins and their associations with cardiovascular disease as described in biomedical reports. This protocol can be easily adapted to study user-selected cellular components and...
2.1K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness03:14

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

1.0K
In this protocol, foundation large language model response quality is improved via augmentation with peer-reviewed, domain-specific scientific articles through a vector embedding mechanism. Additionally, code is provided to aid in performance comparison across large language...
1.0K
The (Spatial) Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition05:15

The (Spatial) Memory Game: Testing the Relationship Between Spatial Language, Object Knowledge, and Spatial Cognition

11.3K
We present a protocol to explore the relationship between spatial language production, spatial memory, and object knowledge. The procedure allows experimental manipulation of, and control over, conditions of object knowledge, language at instruction, and physical location, thus teasing apart cognitive and linguistic models describing interactions between these...
11.3K
Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

8.0K
Existing algorithms generate one solution for a biomarker detection dataset. This protocol demonstrates the existence of multiple similarly effective solutions and presents a user-friendly software to help biomedical researchers investigate their datasets for the proposed challenge. Computer scientists may also provide this feature in their biomarker detection...
8.0K
Retrieval01:12

Retrieval

418
Retrieval is the process of getting information out of memory storage and back into conscious awareness. This ability is essential for daily tasks like brushing hair and teeth, driving to work, and performing job duties. Retrieval occurs in three ways: recall, recognition, and relearning.
Recall involves accessing information without cues, such as during an essay test, where individuals must retrieve facts and concepts from memory unaided. Another example is remembering the name of a colleague...
418

También podría leer

Artículos Relacionados

Artículos vinculados a este trabajo por autores compartidos, revista y gráfico de citas.

Ordenar por
Same author

Desiderata for a biomedical knowledge network: opportunities, challenges and future directions.

Bioinformatics advances·2026
Same author

A case-based explainable graph neural network framework for mechanistic drug repositioning.

Bioinformatics (Oxford, England)·2026
Same author

The NIAID Discovery Portal: a unified search engine for infectious and immune-mediated disease datasets.

mSystems·2025
Same author

Federated Knowledge Retrieval Elevates Large Language Model Performance on Biomedical Benchmarks.

bioRxiv : the preprint server for biology·2025
Same author

Announcing the Biomedical Data Translator: Initial Public Release.

Clinical and translational science·2025
Same author

Drug Repurposing using consilience of Knowledge Graph Completion methods.

bioRxiv : the preprint server for biology·2024
Same journal

NanoporeDB: A Structural Resource Of Multimeric Protein Nanopores For Single-Molecule Sensing.

GigaScience·2026
Same journal

From the Brain Cell Atlas to Precision Neurology: A review of the application of AI-driven multi-omics in brain science.

GigaScience·2026
Same journal

Comparison of Deep Learning Approaches for Extreme Low-SNR Image Restoration.

GigaScience·2026
Same journal

ScopeViewer: A Browser-Based Solution for Visualizing Large Biological Images.

GigaScience·2026
Same journal

ChatMDV: Reducing Technical Barriers in Bioinformatics Analysis using Large Language Models.

GigaScience·2026
Same journal

ClusterGraph: a new tool for visualisation and compression of multidimensional data.

GigaScience·2026
Ver todos los artículos relacionados

Video Experimental Relacionado

Updated: Jan 20, 2026

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

1.3K

Recuperación de Conocimiento Federado Mejora el Rendimiento de Modelos de Lenguaje Grandes en Benchmarks Biomédicos

Janet Joy1, Andrew I Su1

  • 1Department of Integrative Structural and Computational Biology, Scripps Research, La Jolla, CA, USA.

GigaScience
|January 19, 2026
PubMed
Resumen
Este resumen es generado por máquina.

La generación aumentada por recuperación utilizando BioThings Explorer (BTE-RAG) mejora la precisión de los modelos de lenguaje grandes (LLM) en la investigación biomédica. Este marco mejora la corrección fáctica y la exploración mecanística para el descubrimiento de fármacos y la ciencia traslacional.

Palabras clave:
modelos de lenguaje grandesinvestigación biomédicageneración aumentada por recuperaciónexploración mecanísticadescubrimiento de fármacosciencia traslacionalBioThings ExplorerBTE-RAGprecisión fácticaalucinaciones

Más Videos Relacionados

A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
07:35

A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

Published on: October 13, 2023

2.1K
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

1.0K

Videos de Experimentos Relacionados

Last Updated: Jan 20, 2026

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

1.3K
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
07:35

A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

Published on: October 13, 2023

2.1K
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

1.0K

Área de la Ciencia:

  • Investigación biomédica
  • Inteligencia artificial
  • Representación del conocimiento

Sus antecedentes:

  • Los modelos de lenguaje grandes (LLM) ofrecen procesamiento avanzado del lenguaje natural para la investigación biomédica.
  • Los LLM pueden producir imprecisiones fácticas (alucinaciones) debido a la dependencia de datos implícitos.
  • Estas imprecisiones plantean riesgos en aplicaciones biomédicas críticas.

Objetivo del estudio:

  • Desarrollar un marco que mejore la precisión de los LLM en la investigación biomédica.
  • Integrar evidencia mecanística explícita con el razonamiento de los LLM.
  • Mejorar la precisión fáctica y reducir las alucinaciones en las salidas de los LLM.

Principales métodos:

  • Se desarrolló BTE-RAG, un marco de generación aumentado por recuperación.
  • Se integró el razonamiento de LLM con evidencia explícita de BioThings Explorer (federación de API).
  • Se evaluó BTE-RAG frente a métodos solo de LLM en tres conjuntos de datos de benchmark personalizados (mecanismos genéticos, efectos de metabolitos, procesos biológicos de fármacos).

Principales resultados:

  • BTE-RAG mejoró significativamente la precisión en tareas centradas en genes (por ejemplo, la precisión de GPT-4o aumentó del 69,8% al 78,6%).
  • Mejoró la calidad de la respuesta para efectos de metabolitos (por ejemplo, aumento del 82% en alta similitud de coseno para GPT-4o mini).
  • Mejoró la concordancia de respuestas para relaciones fármaco-proceso biológico y superó a modelos alternativos en benchmarks de asociación gen-enfermedad.

Conclusiones:

  • La recuperación de conocimiento federado a través de BTE-RAG ofrece mejoras transparentes de precisión para los LLM.
  • BTE-RAG es una herramienta práctica para la exploración mecanística en la investigación biomédica.
  • El marco apoya la investigación biomédica traslacional al mejorar la confiabilidad de los LLM.