Conferencia
Permanent URI for this collectionhttps://hdl.handle.net/11285/636053
Presentación o disertación realizada dentro de un congreso o evento similar, o como evento académico independiente, tales como: Conferencia inaugural, conferencia magistral, conferencia de clausura.
Browse
Search Results
- Generative artificial intelligence in higher education: a literature mapping perspective(IATED Digital Library, 2024-11-11) García López, Iván Miguel; Ramírez Montoya, María Soledad; Molina Espinosa, José Martín; https://ror.org/03ayjn504Generative Artificial Intelligence (GAI) opens intriguing possibilities for creating unique and innovative content. This systematic literature review (SLR) analysed the published evidence on GAI in higher education institutions from January 2018 to October 2023. Fourteen articles on the topic were identified in the Web of Science (WOS) and Scopus databases. We screened to obtain the most relevant data by applying inclusion, exclusion, and quality criteria. The results identified (1) the characteristics of the publications, (2) methodological trends, and (3) the approaches implemented. The study aims to be of value to the academic community and developers of GAI initiatives
- Design and challenges of open large language model frameworks (Open LLM): a systematic literature mapping(IATED Digital Library, 2024-11-11) García López, Iván Miguel; Ramírez Montoya, María Soledad; Molina Espinosa, José Martín; https://ror.org/03ayjn504Analyzing the frameworks of open large language models (OLLM) is essential to understanding how the management of these artificial intelligence (AI) models can be regulated. This study aims to analyze the evidence published from 2019 to 2024 regarding OLLM frameworks that integrate AI. Systematic mapping was the method for reviewing 227 articles published in the Scopus and Web of Science (WoS) databases. Inclusion, exclusion, and quality criteria filtered the papers to obtain the maximum relevant information. The analysis and classification of articles related to open LLM frameworks and models yielded significant findings per our research questions. The challenges identified were a) improving customization and accuracy through open LLMs, b) latency and efficiency challenges, c) the importance of reliability and security, and d) complex operational management (LLMOps). This review provides a framework for identifying the topic's state of the art and current and emerging research trends.
- Regulatory challenges and optimization strategies for open large language models: a multidimensional framework for efficient management(Springer Link, 2024-10-23) García López, Iván Miguel; Ramírez Montoya, María Soledad; Molina Espinosa, José Martín; https://ror.org/03ayjn504Innovations in artificial intelligence are rapidly transforming various in-dustries, particularly through the development and deployment of Open Large Language Models (OLLMs). However, the absence of a robust regula-tory framework presents significant challenges in ensuring the ethical, safe, and effective use of these models. This research aims to address this gap by proposing a comprehensive regulatory framework designed to optimize the scalability and performance of OLLMs, emphasizing the importance of structured pruning techniques. By integrating both quantitative and qualita-tive analyses, the study will assess the technical capabilities and societal implications of OLLMs, ultimately providing clear guidelines that promote responsible and sustainable innovation. The expected outcomes include the development of a governance model that balances the need for innovation with ethical considerations, offering a pathway for the regulation of OLLMs that supports their continued evolution while safeguarding public interests.
- Modelos de lenguaje grande, la revolución silenciosa en educación, salud e industria: mapeo sistémico de literatura(International Insitute of Systemics, Cybernetics, and Informatics: IIIS, 2024-09-10) García López, Iván Miguel; Ramírez Montoya, María Soledad; Molina Espinosa, José Martín; https://ror.org/03ayjn504Los Modelos de Lenguaje Grande (LLMs) están revolucionando múltiples sectores mediante su capacidad para procesar y generar texto con un alto nivel de coherencia y contexto. Estos modelos no permiten acceso público a su código fuente ni datos de entrenamiento, lo que restringe su uso a la organización que los posee. A diferencia de los modelos de código abierto, los LLMs cerrados presentan tanto oportunidades como desafíos únicos. Nos preguntamos: ¿Cuáles son las estrategias efectivas para integrar LLMs en educación, salud e industria, y cómo pueden manejar los retos éticos, de seguridad y transparencia? El método del estudio es una revisión sistemática de la literatura, analizando artículos de las bases de datos Scopus y Web of Science desde enero de 2019 hasta mayo de 2024. Se seleccionaron los datos más relevantes utilizando criterios de inclusión, exclusión y calidad, y se delimitaron 60 artículos para su análisis. Los hallazgos destacan: (a) oportunidades significativas en la personalización y eficiencia en diversos sectores, (b) marcos de trabajo para la integración efectiva de LLMs y (c) estrategias para abordar desafíos éticos y de seguridad. Este escrito invita a otros investigadores a explorar el uso de LLMs en distintos ámbitos, resaltando su potencial para transformar procesos educativos, médicos e industriales, mientras se asegura un uso responsable y ético.
- How Covid-19 has an impact on formal education: A collective international evaluation of open education in distance learning(14th annual International Conference of Education, Research and Innovation (ICERI2021), 2021-11-09) Stracke, Christian M.; Sharma, R.C.; Swiatek Cassafieres,Cecile; Burgos, Daniel; Bozkurt,Aras; Karakaya, Ozlem; Inamorato dos Santos, Andreia; Mason,Jonathan Charles; Nerantzi, Chrissi; Agbu, Jane-Frances Obiageli; Ossiannilsson, Ebba; Ramírez Montoya, María Soledad; Santos Hermosa, Gema; Shon, Jin Gon; Wan, Marian; Conole, Grainne; Farrow, Robert; https://ror.org/03ayjn504While causing unprecedented disruption worldwide, COVID-19 has also stimulated the mainstreaming of digital technologies in the delivery of formal education. For most key stakeholders – organisations, educators, and students – this has been a new and challenging experience and has been described in policy terms as ‘emergency remote education’. For many students, however, it has either exacerbated or marginalised their opportunity to access formal education. In probing this impact at a deeper level, an international collaboration involving the authors during 2020-2021 focused on reviewing contemporary practices and potentials of open education as a strategic and sustainable response. This paper highlights practices, case studies, and emerging issues from 13 diverse countries, to be globally representative, which include: Australia, Brazil, France, India, Mexico, the Netherlands, Nigeria, Spain, Sweden, South Korea, Taiwan, Turkey, and the United Kingdom. This collection of countries was selected based on researcher contexts and contributions. To date, findings indicate open education has demonstrable benefits for distance learning. More broadly, open educational practices are positioned to shape a ‘new normal’ that embraces ‘global citizenship’ while also being equitable and inclusive. Our aspirations are that such practices will lead to better formal education promoting and ensuring human rights, democracy, lifelong learning, safety, social justice, diversity, cultural sensitivity and inclusivity through strategic and long-term support by all stakeholders in both modes of educational delivery and access: face-to-face and distance learning.