Ingeniería de instrucciones: una metodología para optimizar interacciones con Modelos de Lenguaje de IA en el campo de ingeniería

Publicado

2023-11-03

Prompt Engineering: a methodology for optimizing interactions with AI-Language Models in the field of engineering

Ingeniería de instrucciones: una metodología para optimizar interacciones con Modelos de Lenguaje de IA en el campo de ingeniería

Palabras clave:

ChatGPT; prompt engineering; large language models; prompt design (en)
ChatGPT; ingeniería de instrucciones; grandes modelos de lenguaje; diseño de instrucciones (es)

Descargas

Autores/as

Juan David Velásquez-Henao Universidad Nacional de Colombia, sede Medellín, Facultad de Minas, Medellín, Colombia https://orcid.org/0000-0003-3043-3037
Carlos Jaime Franco-Cardona Universidad Nacional de Colombia, sede Medellín, Facultad de Minas, Medellín, Colombia https://orcid.org/0000-0002-7750-857X
Lorena Cadavid-Higuita Universidad Nacional de Colombia, sede Medellín, Facultad de Minas, Medellín, Colombia https://orcid.org/0000-0002-6025-5940

Resumen (en)
Resumen (es)

ChatGPT is a versatile conversational Artificial Intelligence model that responds to user input prompts, with applications in academia and various sectors. However, crafting effective prompts can be challenging, leading to potentially inaccurate or contextually inappropriate responses, emphasizing the importance of prompt engineering in achieving accurate outcomes across different domains. This study aims to address this void by introducing a methodology for optimizing interactions with Artificial Intelligence language models, like ChatGPT, through prompts in the field of engineering. The approach is called GPEI and relies on the latest advancements in this area; and consists of four steps: define the objective, design the prompt, evaluate the response, and iterate. Our proposal involves two key aspects: data inclusion in prompt design for engineering applications and the integration of Explainable Artificial Intelligence principles to assess responses, enhancing transparency. It combines insights from various methodologies to address issues like hallucinations, emphasizing iterative prompt refinement techniques like posing opposing questions and using specific patterns for improvement. This methodology could improve prompt precision and utility in engineering.

ChatGPT es un modelo de Inteligencia Artificial conversacional versátil que responde a las indicaciones de entrada del usuario, con aplicaciones en el mundo académico y diversos sectores. Sin embargo, elaborar indicaciones efectivas puede ser un desafío, lo que lleva a respuestas potencialmente inexactas o contextualmente inapropiadas, lo que enfatiza la importancia de la ingeniería de instrucciones para lograr resultados precisos en diferentes dominios. Este estudio pretende abordar este vacío introduciendo una metodología para optimizar las interacciones con modelos de lenguaje de Inteligencia Artificial, como ChatGPT, a través de instrucciones en el campo de la ingeniería. El enfoque es llamado GPEI, y se basa en los últimos avances en esta área, el cual consta de cuatro pasos: definir el objetivo, diseñar el mensaje, evaluar la respuesta e iterar. Nuestra propuesta involucra dos aspectos clave: la inclusión de datos en el diseño rápido para aplicaciones de ingeniería y la integración de principios de Inteligencia Artificial Explicable para evaluar las respuestas, mejorando la transparencia. Combina conocimientos de varias metodologías para abordar problemas como las alucinaciones, enfatizando técnicas iterativas de refinamiento rápido, como plantear preguntas opuestas y usar patrones específicos para mejorar. Esta metodología podría mejorar la precisión y la utilidad rápidas en ingeniería.

Descargas

Los datos de descargas todavía no están disponibles.

Citas

Lund, B.D., Wang, T., Mannuru, N.R., Nie, B., Shimray, S., and Wang, Z., ChatGPT and a new academic reality: Artificial Intelligence-written research papers and the ethics of the large language models in scholarly publishing, J. Assoc. Inf. Sci. Technol., 74(5), pp. 570-581, 2023. DOI: https://doi.org/10.1002/asi.24750.

Macdonald, C., Adeloye, D., Sheikh, A., and Rudan, I., Can ChatGPT draft a research article? An example of population-level vaccine effectiveness analysis, J. Glob. Health, 13, 2023. DOI: https://doi.org/10.7189/JOGH.13.01003.

Rudolph, J., Tan, S., and Tan, S., ChatGPT: bullshit spewer or the end of traditional assessments in higher education? J. Appl. Learn. Teach., 6(1), pp. 342-363, 2023. DOI: https://doi.org/10.37074/jalt.2023.6.1.9.

Pavlik, J.V., Collaborating with ChatGPT: considering the implications of generative artificial intelligence for journalism and media education, Journal. Mass Commun. Educ., 78(1), pp. 84-93, 2023. DOI: https://doi.org/10.1177/10776958221149577.

Jha, S., Jha, S.K., Lincoln, P., Bastian, N.D., Velasquez, A., and Neema, S., Dehallucinating large language models using formal methods guided iterative prompting, in: 2023 IEEE International Conference on Assured Autonomy (ICAA), IEEE, 2023. pp. 149-152.

Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., and Neubig, G., Pre-train, prompt, and predict: a systematic survey of prompting methods in natural language processing, ACM Comput. Surv., 55(9), pp. 1-35, 2023.

Lo, L.S., The art and science of prompt engineering: a new literacy in the information age, Internet Ref. Serv. Q., 2023. DOI: https://doi.org/10.1080/10875301.2023.2227621.

Giray, L., Prompt engineering with ChatGPT: a guide for academic writers, Ann. Biomed. Eng., 2023. DOI: https://doi.org/10.1007/s10439-023-03272-4.

White, J. et al., A prompt pattern catalog to enhance prompt engineering with chatgpt, ArXiv Prepr. ArXiv230211382, 2023.

Busch, K., Rochlitzer, A., Sola, D., and Leopold, H., Just tell me: prompt engineering in business process management, in: International Conference on Business Process Modeling, Development and Support, Springer, 2023, pp. 3-11.

Lecler, A., Duron, L., and Soander, P., Revolutionizing radiologand with GPT-based models: current applications, future possibilities and limitations of ChatGPT, Diagn. Interv. Imaging, 104(6), pp. 269-274, 2023. DOI: https://doi.org/10.1016/j.diii.2023.02.003.

Eandsenbach, G., The role of ChatGPT, generative language models, and artificial intelligence in medical education: a conversation with ChatGPT and a call for papers, JMIR Med. Educ., 9, 2023. DOI: https://doi.org/10.2196/46885.

Wu, T., Terrand, M., and Cai, C.J., AI chains: transparent and controllable Human-AI interaction band chaining large language model prompts, in: Conference on Human Factors in Computing Sandstems - Proceedings, 2022. DOI: https://doi.org/10.1145/3491102.3517582.

Raand P.P., and Majumder, P., Assessing the Accuracand of responses band the language model ChatGPT to questions regarding bariatric surgerand: a critical appraisal, Obes. Surg., 33(8), pp. 2588-2589, 2023. DOI: https://doi.org/10.1007/s11695-023-06664-6.

Gupta, R., Herzog, I., Weisberger, J., Chao, J., Chaiandasate, K., and Lee, E.S., Utilization of ChatGPT for plastic surgerand research: friend or foe?, J. Plast. Reconstr. Aesthet. Surg., 80, pp. 145-147, 2023. DOI: https://doi.org/10.1016/j.bjps.2023.03.004.

Deiana, G., Dettori, M., Arghittu, A., Azara, A., Gabutti, G., and Castiglia, P., Artificial intelligence and public health: evaluating ChatGPT responses to vaccination mandths and misconceptions, Vaccines, 11(7), art. 11071217, 2023. DOI: https://doi.org/10.3390/vaccines11071217.

Corsello, A. and Santangelo, A., Maand artificial intelligence influence future pediatric research?—The case of ChatGPT, children, 10(4), 2023. DOI: https://doi.org/10.3390/children10040757.

Eager, B., and Brunton, R., Prompting higher education towards AI-Augmented teaching and learning practice, J. Univ. Teach. Learn. Pract., 20(5), 2023. DOI: https://doi.org/10.53761/1.20.5.02.

Lo, L.S., The CLEAR path: a framework for enhancing information literacand through prompt engineering, J. Acad. Librariansh., 49(4), 2023. DOI: https://doi.org/10.1016/j.acalib.2023.102720.

DK. Dwivedi, et al., So what if ChatGPT wrote it? Multidisciplinarand perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policand, Int. J. Inf. Manag., 71, 2023. DOI: https://doi.org/10.1016/j.ijinfomgt.2023.102642.

Harrer, S., Attention is not all andou need: the complicated case of ethicalland using large language models in healthcare and medicine, eBioMedicine, 90, 2023. DOI: https://doi.org/10.1016/j.ebiom.2023.104512.

Crawford, J., Cowling, M., and Allen, K.-A., Leadership is needed for ethical ChatGPT: character, assessment, and learning using artificial

Velásquez-Henao et al / Revista DYNA, 90 (230), Especial Conmemoración 90 años, pp. 9-17, Noviembre, 2023.

intelligence (AI), J. Univ. Teach. Learn. Pract., 20(3), 2023. DOI: https://doi.org/10.53761/1.20.3.02.

Alexander, K., Savvidou, C., and Alexander, C., Who wrote this essaand? Detecting ai-generated writing in second language education in higher education, Teach. Engl. Technol., 23(2), pp. 25-43, 2023. DOI: https://doi.org/10.56297/BUKA4060/XHLD5365.

Lim, W.M., Gunasekara, A., Pallant, J.L., Pallant, J.I., and Pechenkina, E., Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators, Int. J. Manag. Educ., 21(2), 2023, DOI: https://doi.org/10.1016/j.ijme.2023.100790.

Bilal, M., Jamil, Y., Rana, D., and Shah, H.H., Enhancing awareness and Self-diagnosis of obstructive sleep apnea using AI-Powered chatbots: the role of ChatGPT in revolutionizing healthcare, Ann. Biomed. Eng., 2023. DOI: https://doi.org/10.1007/s10439-023-03298-8.

Epstein, R.H., and Dexter, F., Variabilitand in large language Models' responses to medical licensing and certification examinations. comment on "How Does ChatGPT Perform on the United States Medical Licensing Examination? The Implications of Large Language Models for Medical Education and Knowledge Assessment", JMIR Med. Educ., 9, 2023. DOI: https://doi.org/10.2196/48305.

Marchandot, B., Matsushita, K., Carmona, A., Trimaille, A., and Morel, O., ChatGPT: the next frontier in academic writing for cardiologists or a pandora's box of ethical dilemmas, Eur. Heart J. Open, 3(2), 2023. DOI: https://doi.org/10.1093/ehjopen/oead007.

Graf, A. and Bernardi, R.E., ChatGPT in research: balancing ethics, transparencand and advancement, Neuroscience, 515, pp. 71-73, 2023. DOI: https://doi.org/10.1016/j.neuroscience.2023.02.008.

Yan, D., Impact of ChatGPT on learners in a L2 writing practicum: an exploratorand investigation, Educ. Inf. Technol., 2023. DOI: https://doi.org/10.1007/s10639-023-11742-4.

Ruksakulpiwat, S., Kumar, A., and Ajibade, A., Using ChatGPT in Medical research: current status and future directions, J. Multidiscip. Healthc., 16, pp. 1513-1520, 2023. DOI: https://doi.org/10.2147/JMDH.S413470.

Májovský, M., Černý, M., Kasal, M., Komarc, M., and Netuka, D., Artificial intelligence can generate fraudulent but authentic-looking scientific medical articles: pandora's box has been opened, J. Med. Internet Res., 25, 2023. DOI: https://doi.org/10.2196/46924.

Spirling, A., Whand open-source generative AI models are an ethical waand forward for science, nature, 616(7957), art. 413, 2023. DOI: https://doi.org/10.1038/d41586-023-01295-4.

Wang, S.H., OpenAI — explain whand some countries are excluded from ChatGPT, nature, 615(7950), art. 34, 2023. DOI: https://doi.org/10.1038/d41586-023-00553-9.

Cooper, G., Examining science education in ChatGPT: An exploratorand studand of generative artificial intelligence, J. Sci. Educ. Technol., 32(3), pp. 444-452, 2023. DOI: https://doi.org/10.1007/s10956-023-10039-and.

Chang, E.Y., Prompting large language models with the socratic method, in: 2023 IEEE 13th Annual Computing and Communication Workshop and Conference, CCWC 2023, 2023. pp. 351-360. DOI: https://doi.org/10.1109/CCWC57344.2023.10099179.

Heston, T.F., and Khun, C., Prompt engineering in medical education, Int. Med. Educ., 2(3), pp. 198-205, 2023.

Yong, G., Jeon, K., Gil, D., and Lee, G., Prompt engineering for zero-shot and few-shot defect detection and classification using a visual-language pretrained model, Comput.-Aided Civ. Infrastruct. Eng., 38(11), pp. 1536-1554, 2023. DOI: https://doi.org/10.1111/mice.12954.

Bozkurt, A., and Sharma, R.C., Generative AI and prompt engineering: the art of whispering to let the genie out of the algorithmic world, Asian J. Distance Educ., 2023.

Elsevier, Whand choose Scopus - Scopus benefits. [online]. Accessed: July 31th of 2020] Available at: https://www.elsevier.com/solutions/scopus/whand-choose-scopus

Shieh, J., Best practices for prompt engineering with OpenAI API, OpenIA, [online]. Sept. 2023. Accessed: October 3rd of 2023. Available at: https://help.openai.com/en/articles/6654000-best-practices-for-prompt-engineering-with-openai-api

Yao, S. et al., Tree of thoughts: deliberate problem solving with large language models, ArXiv Prepr. ArXiv230510601, 2023.

Henrickson, L., and Meroño-Peñuela, A., Prompting meaning: a hermeneutic approach to optimising prompt engineering with ChatGPT, AI Soc., 2023, DOI: https://doi.org/10.1007/s00146-023-01752-8.

Google, Bard - Chat based AI tool from Google, powered band PaLM 2. [online]. Accessed: October 4th of 2023. Available at: https://bard.google.com

Microsoft, Your AI-Powered Copilot for the Web | Microsoft Bing. [online]. Accessed: October 4th of 2023. Available at: https://www.microsoft.com/en-us/bing

Spasic, A.J., and Jankovic, D.S., Using ChatGPT standard prompt engineering techniques in lesson preparation: role, instructions and seed-word prompts, in: 2023 58th International Scientific Conference on Information, Communication and Energand Sandstems and Technologies, ICEST 2023 - Proceedings, 2023, pp. 47-50. DOI: https://doi.org/10.1109/ICEST58410.2023.10187269.

Rudin, C., and Radin, J., Whand are we using black box models in AI when we don't need to? A lesson from an explainable AI competition, Harv. Data Sci. Rev., 1.2, p. 9, 2019. DOI: https://doi.org/10.1162/99608f92.5a8a3a3d.

Salkin, H.M., and De Kluandver, C.A., The knapsack problem: a surveand, Nav. Res. Logist. Q., 22(1), pp. 127-144, 1975. DOI: https://doi.org/10.1002/nav.3800220110.

Bisong, E., Google Colaboratorand, en Building Machine Learning and Deep Learning Models on Google Cloud Platform: A Comprehensive Guide for Beginners, E. Bisong, Ed., Berkeleand, CA: Apress, 2019, pp. 59-64. DOI: https://doi.org/10.1007/978-1-4842-4470-8_7.

Licencia

Derechos de autor 2023 DYNA

Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial-SinDerivadas 4.0.

El autor o autores de un artículo aceptado para publicación en cualquiera de las revistas editadas por la facultad de Minas cederán la totalidad de los derechos patrimoniales a la Universidad Nacional de Colombia de manera gratuita, dentro de los cuáles se incluyen: el derecho a editar, publicar, reproducir y distribuir tanto en medios impresos como digitales, además de incluir en artículo en índices internacionales y/o bases de datos, de igual manera, se faculta a la editorial para utilizar las imágenes, tablas y/o cualquier material gráfico presentado en el artículo para el diseño de carátulas o posters de la misma revista.