|
The maturity of Brazilian companies on the adoption of industry 4.0 practices Maria Amelia Eliseo, Ismar Frango Silveira, Valéria Farinazzo Martins, Cibelle Albuquerque de la Higuera Amato, Daniela Vieira, Leonildo Carnevalli Junior, Gabriel Batista Cristiano, Felipe Chuang, Fabio Tanikawa |
51-68 |
|
Distilling ChatGPT for Short Story Generation Adrian Neață, Ştefan Trăuşan-Matu |
69-81 |
Universidade Presbiteriana Mackenzie
São Paulo SP 01239-001, Brazil
Abstract. The fourth industrial revolution, commonly known as Industry 4.0, has brought significant transformations to the global production model by placing information technologies at its core. This technological trend and ecosystem encompass various concepts aimed at optimizing industrial processes, including enhanced resource optimization and flexibility through the integration of human factors and technology. However, the benefits of this model are not uniformly distributed among countries, and peripheral or semi-peripheral countries, within a capitalist framework, face a competitive disadvantage due to limited access to technology when compared to developed nations. In this article, we examine the current state of Industry 4.0 development in Brazil by focusing on the adoption of digital transformation technologies and processes within domestic companies. Our analysis reveals that Brazil is still in the early stages of maturity regarding the implementation of Industry 4.0 practices, and there is a scarcity of research on this topic within the national context.
Keywords: Industry 4.0, Brazil, Digital transformation
Cite this paper as:
Eliseo, M. A., Silveira, I. F., Farinazzo Martins, V., Albuquerque de la Higuera Amato, C., Vieira, D., Carnevalli Junior, L., Cristiano, G. B., Chuang, F., Tanikawa, F. The maturity of Brazilian companies on the adoption of industry 4.0 practices.
International Journal of User-System Interaction 15(3),
51-68, 2022.
1 University Politehnica of Bucharest
313 Splaiul Independentei, Bucharest, Romania
2 Institutul de Cercetări în Inteligenţa Artificială
Calea 13 Septembrie nr. 13, Bucureşti
3 Academy of Romanian Scientists
Splaiul Indpendentei 54, Bucharest, Romania
Abstract. The paper presents an analysis of the knowledge distillation process, encompassing the preparation of a ChatGPT-generated story dataset, the finetuning of a small GPT-2 model, and the evaluation of short story quality and coherence. Through extensive experiments and evaluations, state-of-the-art results are obtained when compared to models of similar sizes (between 100M and 400M parameters) and we believe our approach can boost the performance of NLG models in general. Furthermore, an impromptu examination of the texts generated by ChatGPT reveals biases towards certain words and phrases.
Keywords: natural language generation; knowledge distillation; story generation; ChatGPT; GPT2
Cite this paper as:
Neață, A., Trăuşan-Matu, S. Distilling ChatGPT for Short Story Generation.
International Journal of User-System Interaction 15(3),
69-81, 2022.