Advancing Life Cycle Assessment of Sustainable Green Hydrogen Production Using Domain-Specific Fine-Tuning by Large Language Models Augmentation DOI Creative Commons

Yajing Chen,

Urs Liebau,

Shreyas Mysore Guruprasad

et al.

Machine Learning and Knowledge Extraction, Journal Year: 2024, Volume and Issue: 6(4), P. 2494 - 2514

Published: Nov. 4, 2024

Assessing the sustainable development of green hydrogen and assessing its potential environmental impacts using Life Cycle Assessment is crucial. Challenges in LCA, like missing data, are often addressed machine learning, such as artificial neural networks. However, to find an ML solution, researchers need read extensive literature or consult experts. This research demonstrates how customised LLMs, trained with domain-specific papers, can help overcome these challenges. By starting small by consolidating papers focused on LCA proton exchange membrane water electrolysis, which produces hydrogen, applications LCA. These uploaded OpenAI create LlamaIndex, enabling future queries. Using LangChain framework, query model (GPT-3.5-turbo), receiving tailored responses. The results demonstrate that LLMs assist providing suitable solutions address data inaccuracies gaps. ability quickly LLM receive integrated response across relevant sources presents improvement over manually retrieving reading individual papers. shows leveraging fine-tuned empower conduct LCAs more efficiently effectively.

Language: Английский

Analyzing and Mitigating Cultural Hallucinations of Commercial Language Models in Turkish DOI Creative Commons
Yiğithan Boztemir, Nilüfer Çalışkan

Published: May 7, 2024

In an era where artificial intelligence is increasingly interfacing with diverse cultural contexts, the ability of language models to accurately represent and adapt these contexts paramount importance.The present research undertakes a meticulous evaluation three prominent commercial models-Google Gemini 1.5, ChatGPT-4, Anthropic's Claude 3 Sonet-with focus on their handling Turkish language.Through dual approach quantitative metrics, Cultural Inaccuracy Score (CIS) Sensitivity Index (CSI), alongside qualitative analyses via detailed case studies, disparities in model performances were highlighted.Notably, Sonet exhibited superior sensitivity, underscoring effectiveness its advanced training methodologies.Further analysis revealed that all demonstrated varying degrees competence, suggesting significant room for improvement.The findings emphasize necessity enriched diversified datasets, innovative algorithmic enhancements, reduce inaccuracies enhance models' global applicability.Strategies mitigating hallucinations are discussed, focusing refinement processes continuous foster improvements AI adaptiveness.The study aims contribute ongoing technologies, ensuring they respect reflect rich tapestry human cultures.

Language: Английский

Citations

24
Evolution and Optimization of Language Model Architectures: From Foundations to Future Directions DOI

Zainab M. AlQenaei

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 233 - 249

Published: Jan. 1, 2025

Language: Английский

Citations

6
The sociolinguistic foundations of language modeling DOI Creative Commons
Jack Grieve,

Sara Bartl,

Matteo Fuoli

et al.

Frontiers in Artificial Intelligence, Journal Year: 2025, Volume and Issue: 7

Published: Jan. 13, 2025

In this article, we introduce a sociolinguistic perspective on language modeling. We claim that models in general are inherently modeling varieties of , and consider how insight can inform the development deployment models. begin by presenting technical definition concept variety as developed sociolinguistics. then discuss could help us better understand five basic challenges modeling: social bias, domain adaptation, alignment, change scale . argue to maximize performance societal value it is important carefully compile training corpora accurately represent specific being modeled, drawing theories, methods, descriptions from field

Language: Английский

Citations

2
Comparative Evaluation of Commercial Large Language Models on PromptBench: An English and Chinese Perspective DOI Creative Commons
Shiyu Wang, Qian Ouyang, Bing Wang

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 27, 2024

Abstract This study embarks on an exploration of the performance disparities observed between English and Chinese in large language models (LLMs), motivated by growing need for multilingual capabilities artificial intelligence systems. Utilizing a comprehensive methodology that includes quantitative analysis model outputs qualitative assessment nuances, research investigates underlying reasons these discrepancies. The findings reveal significant variations LLMs across two languages, with pronounced challenge accurately processing generating text Chinese. gap underscores limitations current handling complexities inherent languages distinct grammatical structures cultural contexts. implications this are far-reaching, suggesting critical development more robust inclusive can better accommodate linguistic diversity. entails not only enrichment training datasets wider array but also refinement architectures to grasp subtleties different Ultimately, contributes ongoing discourse enhancing LLMs, aiming pave way equitable effective tools cater global user base.

Language: Английский

Citations

15
Generative AI, Research Ethics, and Higher Education Research: Insights from a Scientometric Analysis DOI Creative Commons
Saba Qadhi, Ahmed Alduais, Youmen Chaaban

et al.

Information, Journal Year: 2024, Volume and Issue: 15(6), P. 325 - 325

Published: June 2, 2024

In the digital age, intersection of artificial intelligence (AI) and higher education (HE) poses novel ethical considerations, necessitating a comprehensive exploration this multifaceted relationship. This study aims to quantify characterize current research trends critically assess discourse on AI applications within HE. Employing mixed-methods design, we integrated quantitative data from Web Science, Scopus, Lens databases with qualitative insights selected studies perform scientometric content analyses, yielding nuanced landscape utilization in Our results identified vital areas through citation bursts, keyword co-occurrence, thematic clusters. We provided conceptual model for integration HE, encapsulating dichotomous perspectives AI’s role education. Three clusters were identified: frameworks policy development, academic integrity creation, student interaction AI. The concludes that, while offers substantial benefits educational advancement, it also brings challenges that necessitate vigilant governance uphold standards. implications extend policymakers, educators, developers, highlighting need guidelines, literacy, human-centered tools.

Language: Английский

Citations

12
Will Artificial Intelligence Affect How Cultural Heritage Will Be Managed in the Future? Responses Generated by Four genAI Models DOI Creative Commons
Dirk Spennemann

Heritage, Journal Year: 2024, Volume and Issue: 7(3), P. 1453 - 1471

Published: March 11, 2024

Generative artificial intelligence (genAI) language models have become firmly embedded in public consciousness. Their abilities to extract and summarise information from a wide range of sources their training data attracted the attention many scholars. This paper examines how four genAI large (ChatGPT, GPT4, DeepAI, Google Bard) responded prompts, asking (i) whether would affect cultural heritage will be managed future (with examples requested) (ii) what dangers might emerge when relying heavily on guide professionals actions. The systems provided examples, commonly drawing extending status quo. Without doubt, AI tools revolutionise execution repetitive mundane tasks, such as classification some classes artifacts, or allow for predictive modelling decay objects. Important were used assess purported power extract, aggregate, synthesize volumes multiple sources, well ability recognise patterns connections that people may miss. An inherent risk ‘results’ presented by is are ‘artifacts’ system rather than being genuine. Since present unable purposively generate creative innovative thoughts, it left reader determine any text out ordinary meaningful nonsensical. Additional risks identified use without required level literacy overreliance lead deskilling general practitioners.

Language: Английский

Citations

9
Reinforcement Learning for Optimized Information Retrieval in LLaMA DOI Creative Commons
Chien-Hung Tu, Hsien-Jung Hsu, Shih-Wen Chen

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 10, 2024

Abstract This study introduces a new approach to enhance information retrieval accuracy in Large Language Models (LLMs) by integrating specially designed reinforcement learning algorithm into the LLaMA model. The research focuses on developing and implementing an that dynamically adapts model's response strategies user queries, based combination of dynamical systems theory relativistic physics. Empirical results demonstrate Optimized model exhibits significant improvements accuracy, relevance, coherence across various tasks compared Baseline LLaMA. advancement not only showcases potential realm natural language processing but also marks considerable step forward development AI capable nuanced understanding decision-making. study's findings have profound implications for future research, particularly enhancing practical applicability LLMs complex, real-world scenarios, set benchmark integration machine techniques models.

Language: Английский

Citations

6
Generative AI in Education: Technical Foundations, Applications, and Challenges DOI Creative Commons
Sheikh Faisal Rashid, Nghia Duong‐Trung,

Niels Pinkwart

et al.

Artificial intelligence, Journal Year: 2024, Volume and Issue: unknown

Published: May 20, 2024

Generative artificial intelligence (AI) (GenAI) has emerged as a transformative force in various fields, and its potential impact on education is particularly profound. This chapter presents the development trends of “GenAI Education” by exploring technical background, diverse applications, multifaceted challenges associated with adoption education. The briefly introduces background GenAI, large language models (LLMs) such ChatGPT & Co. It provides key concepts, models, recent technological advances. then navigates through applications GenAI or LLMs education, examining their different levels including school, university, vocational training. will highlight how reshaping educational landscape real-world examples case studies, from personalized learning experiences to content creation assessment. also discusses technical, ethical, organizational/educational using technology

Language: Английский

Citations

6
Expansive data, extensive model: Investigating discussion topics around LLM through unsupervised machine learning in academic papers and news DOI Creative Commons

Hae Sun Jung,

Haein Lee,

Young Seok Woo

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0304680 - e0304680

Published: May 31, 2024

This study presents a comprehensive exploration of topic modeling methods tailored for large language model (LLM) using data obtained from Web Science and LexisNexis June 1, 2020, to December 31, 2023. The collection process involved queries focusing on LLMs, including “Large model,” “LLM,” “ChatGPT.” Various approaches were evaluated based performance metrics, diversity coherence. latent Dirichlet allocation (LDA), nonnegative matrix factorization (NMF), combined models (CTM), bidirectional encoder representations Transformers (BERTopic) employed evaluation. Evaluation metrics computed across platforms, with BERTopic demonstrating superior in coherence both Science. experiment result reveals that news articles maintain balanced coverage various topics mainly focus efforts utilize LLM specialized domains. Conversely, research papers are more concise concentrated the technology itself, emphasizing technical aspects. Through insights gained this study, it becomes possible investigate future path challenges LLMs should tackle. Additionally, they could offer considerable value enterprises deliver services.

Language: Английский

Citations

4
A Comparative Analysis of Cultural Alignment in Large Language Models in Bilingual Contexts DOI Open Access

Ximen Yuan,

Jinshan Hu, Qian Zhang

et al.

Published: June 10, 2024

Artificial intelligence (AI) systems, particularly those capable of natural language processing, are increasingly becoming integral to diverse aspects human life and interaction. Understanding the cultural biases embedded within AI, especially in how it aligns with specific values, is crucial for ensuring its effective equitable deployment. This research examines alignment AI-generated responses mainstream Chinese such as Confucian harmony, Daoist balance, collectivism, respect authority, family-centric principles. By analyzing both English, study highlights discrepancies inherent AI offering valuable insights into their implications development. The findings reveal that while demonstrates general significant variations exist between contexts, emphasizing importance linguistic specificity interactions. Quantitative metrics thematic analyses demonstrate necessity culturally aware contributing broader discourse on ethical development providing guidance creating more inclusive adaptable systems.

Language: Английский

Citations

4