A Scoping Review of Large Language Models: Architecture and Applications DOI
Hicham Moujahid,

Karima Boutahar,

Oussama El Gannour

и другие.

Опубликована: Май 16, 2024

Large Language Models (LLMs) have recently exhibited impressive usability of natural language processing to perform different outcomes. A plethora research contributions encompassing a wide range areas, including architectural advances, better training strategies, context length enhancements, fine-tuning, multimodal LLMs, robotics, datasets, benchmarking, efficiency, and more, been made possible by this accomplishment. It becomes harder understand the overall progress in LLM when methods keep changing at quick pace discoveries become commonplace. As methodologies continue improve more common research, it has increasingly difficult field. concise but thorough summary recent advances is essential for scientific community access to, considering wealth literature on LLMs. Our careful autonomous review explores innovative themes forefront while also diving into pertinent prior concepts.

Язык: Английский

Multimodal Large Language Models: A Survey DOI
Jiayang Wu, Wensheng Gan, Zefeng Chen

и другие.

2021 IEEE International Conference on Big Data (Big Data), Год журнала: 2023, Номер unknown

Опубликована: Дек. 15, 2023

The exploration of multimodal language models integrates multiple data types, such as images, text, language, audio, and other heterogeneity. While the latest large excel in text-based tasks, they often struggle to understand process types. Multimodal address this limitation by combining various modalities, enabling a more comprehensive understanding diverse data. This paper begins defining concept examining historical development algorithms. Furthermore, we introduce range products, focusing on efforts major technology companies. A practical guide is provided, offering insights into technical aspects models. Moreover, present compilation algorithms commonly used datasets, providing researchers with valuable resources for experimentation evaluation. Lastly, explore applications discuss challenges associated their development. By addressing these aspects, aims facilitate deeper potentiality domains.

Язык: Английский

Процитировано

46
A review on cultivating effective learning: synthesizing educational theories and virtual reality for enhanced educational experiences DOI Creative Commons

Fatma Mallek,

Tehseen Mazhar,

Syed Faisal Abbas Shah

и другие.

PeerJ Computer Science, Год журнала: 2024, Номер 10, С. e2000 - e2000

Опубликована: Май 23, 2024

Immersive technology, especially virtual reality (VR), transforms education. It offers immersive and interactive learning experiences. This study presents a systematic review focusing on VR’s integration with educational theories in higher The evaluates the literature VR applications combined pedagogical frameworks. aims to identify effective strategies for enhancing experiences through VR. process involved analyzing studies about theories, methodologies, outcomes, effectiveness. Findings show that improves outcomes when aligned such as constructivism, experiential learning, collaborative learning. These integrations offer personalized, immersive, highlights importance of incorporating principles into application development. suggests promising direction future research implementation approach maximize value, across settings.

Язык: Английский

Процитировано

16
Can Large Language Models Aid Caregivers of Pediatric Cancer Patients in Information Seeking? A Cross‐Sectional Investigation DOI Creative Commons
Emre Sezgın, D Jackson, A. Baki Kocaballı

и другие.

Cancer Medicine, Год журнала: 2025, Номер 14(1)

Опубликована: Янв. 1, 2025

ABSTRACT Purpose Caregivers in pediatric oncology need accurate and understandable information about their child's condition, treatment, side effects. This study assesses the performance of publicly accessible large language model (LLM)‐supported tools providing valuable reliable to caregivers children with cancer. Methods In this cross‐sectional study, we evaluated four LLM‐supported tools—ChatGPT (GPT‐4), Google Bard (Gemini Pro), Microsoft Bing Chat, SGE—against a set frequently asked questions (FAQs) derived from Children's Oncology Group Family Handbook expert input (In total, 26 FAQs 104 generated responses). Five experts assessed LLM responses using measures including accuracy, clarity, inclusivity, completeness, clinical utility, overall rating. Additionally, content quality was readability, AI disclosure, source credibility, resource matching, originality. We used descriptive analysis statistical tests Shapiro–Wilk, Levene's, Kruskal–Wallis H ‐tests, Dunn's post hoc for pairwise comparisons. Results ChatGPT shows high when by experts. also performed well, especially accuracy clarity responses, whereas Chat SGE had lower scores. Regarding disclosure being AI, it observed less which may have affected maintained balance between response clarity. most readable answered complexity. varied significantly ( p < 0.001) across all evaluations except inclusivity. Through our thematic free‐text comments, emotional tone empathy emerged as unique theme mixed feedback on expectations be empathetic. Conclusion can enhance caregivers' knowledge oncology. Each has strengths areas improvement, indicating careful selection based specific contexts. Further research is required explore application other medical specialties patient demographics, assessing broader applicability long‐term impacts.

Язык: Английский

Процитировано

2
Large Language Models in Education: Vision and Opportunities DOI
Wensheng Gan,

Zhenlian Qi,

Jiayang Wu

и другие.

2021 IEEE International Conference on Big Data (Big Data), Год журнала: 2023, Номер unknown

Опубликована: Дек. 15, 2023

With the rapid development of artificial intelligence technology, large language models (LLMs) have become a hot research topic. Education plays an important role in human social and progress. Traditional education faces challenges such as individual student differences, insufficient allocation teaching resources, assessment effectiveness. Therefore, applications LLMs field digital/smart broad prospects. The on educational (EduLLMs) is constantly evolving, providing new methods approaches to achieve personalized learning, intelligent tutoring, goals, thereby improving quality learning experience. This article aims investigate summarize application smart education. It first introduces background motivation explains essence LLMs. then discusses relationship between digital EduLLMs summarizes current status models. main contributions are systematic summary vision background, motivation, for (LLM4Edu). By reviewing existing research, this provides guidance insights educators, researchers, policy-makers gain deep understanding potential LLM4Edu. further advancing LLM4Edu, while still facing technical, ethical, practical requiring exploration.

Язык: Английский

Процитировано

39
Enhancing E-Learning Experience Through Embodied AI Tutors in Immersive Virtual Environments: A Multifaceted Approach for Personalized Educational Adaptation DOI
Fatemeh Sarshartehrani, Elham Mohammadrezaei, Majid Behravan

и другие.

Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 272 - 287

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

7
Large language models and medical education: a paradigm shift in educator roles DOI Creative Commons
Li Zhui, Fenghe Li, Qining Fu

и другие.

Smart Learning Environments, Год журнала: 2024, Номер 11(1)

Опубликована: Июнь 5, 2024

Abstract This article meticulously examines the transformation of educator roles in medical education against backdrop emerging large language models (LLMs). Traditionally, educators have played a crucial role transmitting knowledge, training skills, and evaluating educational outcomes. However, advent LLMs such as Chat Generative Pre-trained Transformer-4 has expanded enriched these traditional by leveraging opportunities to enhance teaching efficiency, foster personalised learning, optimise resource allocation. imbued with new connotations. Concurrently, present challenges education, ensuring accuracy information, reducing bias, minimizing student over-reliance, preventing patient privacy exposure safeguarding data security, enhancing cultivation empathy, maintaining academic integrity. In response, are called adopt including experts information management, navigators guardians integrity, defenders clinical practice. The emphasises connotations attributes teacher's role, underscoring their irreplaceable value AI-driven evolution education. Educators portrayed not just users advanced technology, but also custodians essence

Язык: Английский

Процитировано

6
Large language model to multimodal large language model: A journey to shape the biological macromolecules to biological sciences and medicine DOI Creative Commons
Manojit Bhattacharya, Soumen Pal, Srijan Chatterjee

и другие.

Molecular Therapy — Nucleic Acids, Год журнала: 2024, Номер 35(3), С. 102255 - 102255

Опубликована: Июнь 15, 2024

After ChatGPT was released, large language models (LLMs) became more popular. Academicians use or LLM for different purposes, and the of is increasing from medical science to diversified areas. Recently, multimodal (MLLM) has also become Therefore, we comprehensively illustrate MLLM a complete understanding. We aim simple extended reviews LLMs MLLMs broad category readers, such as researchers, students in fields, other academicians. The review article illustrates models, their working principles, applications fields. First, demonstrate technical concept LLMs, principle, Black Box, evolution LLMs. To explain discuss tokenization process, token representation, relationships. extensively application biological macromolecules, science, MLLMs. Finally, limitations, challenges, future prospects acts booster dose clinicians, primer molecular biologists, catalyst scientists, benefits

Язык: Английский

Процитировано

6
Revolutionizing Higher Education: Unleashing the Potential of Large Language Models for Strategic Transformation DOI Creative Commons
Mohamed Diab Idris, X.D. Feng, Vladimir Dyo

и другие.

IEEE Access, Год журнала: 2024, Номер 12, С. 67738 - 67757

Опубликована: Янв. 1, 2024

This paper investigates the transformative potential of Large Language Models (LLMs) within higher education, highlighting their capacity to reshape academic landscape. By examining complex impact LLMs across critical areas Higher Education Institutions (HEIs), including role HEIs as gatekeepers knowledge, providers credentials, research centres, incubators innovation, drivers social change and employers. In addition integrity, future intellectual property, public perception. The findings this indicate that can empower transformation in by revolutionising various aspects academia. aim is unveil profound implications integrating these cutting-edge technologies. comprehensive study reveals significant impacts challenges associated with using settings, which achieved through a detailed analysis current literature. core suggest hold promise trigger advancements education. also discusses innovative LLMs, it outlines path for effective use HEIs, emphasising importance thoughtful approach maximise educational benefits. must address thoughtfully, ensuring integration aligns fundamental objectives promoting thinking, personal growth.

Язык: Английский

Процитировано

5
Large Language Models as Evaluators in Education: Verification of Feedback Consistency and Accuracy DOI Creative Commons
Hyein Seo, Taewook Hwang, Jeesu Jung

и другие.

Applied Sciences, Год журнала: 2025, Номер 15(2), С. 671 - 671

Опубликована: Янв. 11, 2025

The recent advancements in large language models (LLMs) have brought significant changes to the field of education, particularly generation and evaluation feedback. LLMs are transforming education by streamlining tasks like content creation, feedback generation, assessment, reducing teachers’ workload improving online efficiency. This study aimed verify consistency reliability as evaluators conducting automated evaluations using various based on five educational criteria. analysis revealed that while were capable performing consistent under certain conditions, a lack was observed both among across for other Notably, low agreement human correlated with reduced LLM evaluations. Furthermore, variations results influenced factors such prompt strategies model architecture, highlighting complexity achieving reliable assessments LLMs. These findings suggest potential transform systems, careful selection combination essential improve their align performance settings.

Язык: Английский

Процитировано

0
GRADERS OF THE FUTURE: COMPARING THE CONSISTENCY AND ACCURACY OF GPT4 AND PRE-SERVICE TEACHERS IN PHYSICS ESSAY QUESTION ASSESSMENTS DOI Open Access

XU Yu-bin,

Lin Liu, Jianwen Xiong

и другие.

Journal of Baltic Science Education, Год журнала: 2025, Номер 24(1), С. 187 - 207

Опубликована: Фев. 25, 2025

As the development and application of large language models (LLMs) in physics education progress, well-known AI-based chatbot ChatGPT4 has presented numerous opportunities for educational assessment. Investigating potential AI tools practical assessment carries profound significance. This study explored comparative performance human graders scoring upper-secondary essay questions. Eighty students’ responses to two questions were evaluated by 30 pre-service teachers ChatGPT4. The analysis highlighted their consistency accuracy, including intra-human comparisons, GPT grading at different times, human-GPT variations across cognitive categories. intraclass correlation coefficient (ICC) was used assess consistency, while accuracy illustrated through Pearson with expert scores. findings reveal that demonstrated higher scoring, scorers showed superior most instances. These results underscore strengths limitations using LLMs assessments. high can be valuable standardizing assessments diverse contexts, nuanced understanding flexibility are irreplaceable handling complex subjective evaluations. Keywords: Physics question assessment, grader, Human graders.

Язык: Английский

Процитировано

0