The core technology behind and beyond ChatGPT: A comprehensive review of language models in educational research

• Leong, Kelvin ^[1] ; Sung, Anna ^[1] ; Jones, Lewis ^[1]
  1. [1] University of Chester

     University of Chester

     Reino Unido

     
• Localización: IJERI: International journal of Educational Research and Innovation, ISSN-e 2386-4303, n. 20, 2023
• Idioma: inglés
• DOI: 10.46661/ijeri.8449
• Títulos paralelos:
  • La tecnología central detrás y más allá de ChatGPT: Una revisión exhaustiva de los modelos de lenguaje en la investigación educativa
• Enlaces
  • Texto completo
• Resumen
  • español

    ChatGPT ha atraído una gran atención en el sector educativo. Dado que la tecnología central detrás de ChatGPT es el modelo de lenguaje, este estudio tiene como objetivo revisar críticamente publicaciones relacionadas y sugerir la dirección futura del modelo de lenguaje en la investigación educativa. Nuestro objetivo es abordar tres preguntas: i) cuál es la tecnología central detrás de ChatGPT, ii) cuál es el nivel de conocimiento de la investigación relacionada y iii) la dirección del potencial de investigación. Se llevó a cabo una revisión crítica de publicaciones relacionadas con el fin de evaluar el estado actual del conocimiento del modelo lingüístico en la investigación educativa. Además, sugerimos un marco rector para futuras investigaciones sobre modelos lingüísticos en educación. Nuestro estudio respondió rápidamente a las preocupaciones planteadas por el uso de ChatGPT en la industria educativa y proporciona a la industria una descripción general completa y sistemática de las tecnologías relacionadas. Creemos que esta es la primera vez que se realiza un estudio para revisar sistemáticamente el nivel de conocimiento del modelo lingüístico en la investigación educativa.

  • English

    ChatGPT has garnered significant attention within the education industry. Given the core technology behind ChatGPT is language model, this study aims to critically review related publications and suggest future direction of language model in educational research. We aim to address three questions: i) what is the core technology behind ChatGPT, ii) what is the state of knowledge of related research and iii) the potential research direction. A critical review of related publications was conducted in order to evaluate the current state of knowledge of language model in educational research. In addition, we further suggest a purpose oriented guiding framework for future research of language model in education. Our study promptly responded to the concerns raised by ChatGPT from the education industry and offers the industry with a comprehensive and systematic overview of related technologies. We believe this is the first time that a study has been conducted to systematically review the state of knowledge of language model in educational research. 

• Referencias bibliográficas
  • Ba, S., Hu, X., Stein, D. & Liu, Q. (2023). Assessing cognitive presence in online inquiry-based discussion through text classification...
  • Baron, N. (2023). Even kids are worried ChatGPT will make them lazy plagiarists, says a linguist who studies tech’s effect on reading, writing...
  • Bengio, Y. & Senecal, J.S. (2008). Adaptive importance sampling to accelerate training of a neural probabilistic language model. IEEE...
  • Beseiso, M., Alzubi, O.A. & Rashaideh, H. (2021). A novel automated essay scoring approach for reliable higher educational assessments....
  • Botarleanu, R.M., Dascalu, M., Allen, L.K., Crossley, S.A. & McNamara, D.S. (2021). Automated Summary Scoring with ReaderBench. In A....
  • Condor, A. (2020). Exploring automatic short answer grading as a tool to assist in human rating. In I. Bittencourt, M. Cukurova, K. Muldner,...
  • Dempsey, J. (2023). AI: Arguing its Place in Higher Education. Higher Education Digest. https://www.highereducationdigest.com/ai-arguing-its-place-in-higher-education/
  • Devlin, J., Chang, M.W., Lee, K. & Toutanova, K. (2018). Bert: pre-training of deep bidirectional transformers for language understanding....
  • Dimzon, F.D. & Pascual, R.M. (2020). An automatic phoneme recognizer for children’s filipino read speech. 2020 IEEE International Conference...
  • Van-Dis, E.A.M., Bollen, J., Zuidema, W., Van-Rooij, R. and Bockting, C.L. (2023). ChatGPT: five priorities for research. Nature, 614(7947),...
  • Du, H., Xing, W. & Pei, B. (2021). Automatic text generation using deep learning: providing large-scale support for online learning communities....
  • Dyulicheva, Y.Y. (2021). Learning Analytics in MOOCS as an Instrument for Measuring Math Anxiety. Voprosy Obrazovaniya-Educational Studies...
  • Esmaeilzadeh, S., Williams, B., Shamsi, D. & Vikingstad, O. (2022). Providing insights for open-response surveys via end-to-end context-aware...
  • Fernandez, N., Ghosh, A., Liu, N., Wang, Z., Choffin, B., Baraniuk, R. & Lan, A. (2022). Automated scoring for reading comprehension via...
  • Firoozi, T., Mohammadi, H. & Gierl, M.J. (2022). Using active learning methods to strategically select essays for automated scoring. Educational...
  • Fitzpatrick, D. (2023). Overcoming ChatGPT fear in 3 steps. FE News. https://www.fenews.co.uk/exclusive/overcoming-chatgpt-fear-in-3-steps/
  • Geller, S.A., Gal, K., Segal, A., Sripathi, K., Kim, H.G., Facciotti, M.T., Igo, M., et al. (2021). New methods for confusion detection in...
  • Gift, T. & Norman, J. (2023). AI makes university honour codes more necessary than ever. Times Higher Education (THE). https://www.timeshighereducation.com/blog/ai-makes-university-honour-codes-more-necessary-ever
  • Goel, V., Sahnan, D., Venktesh, V., Sharma, G., Dwivedi, D. & Mohania, M. (2022). K-12BERT: BERT for K-12 education. In M. Rodrigo, N....
  • Goldberg, Y. & Levy, O. (2014). Word2vec Explained: deriving Mikolov et al.’s negative-sampling word-embedding method. arXi. https://doi.org/10.48550/ARXIV.1402.3722
  • Greenhouse, S. (2023). US experts warn AI likely to kill off jobs – and widen wealth inequality. The Guardian. https://www.theguardian.com/technology/2023/feb/08/ai-chatgpt-jobs-economy-inequality
  • Han, X., Zhang, Z., Ding, N., Gu, Y., Liu, X., Huo, Y., Qiu, J., et al. (2021). Pre-trained models: Past, present and future. AI Open, 2,...
  • Hao, Y., Li, H., Ding, W., Wu, Z., Tang, J., Luckin, R. & Liu, Z. (2021). Multi-task learning based online dialogic instruction detection...
  • Hess, F. (2023). Will ChatGPT Be A Blow To Learning, Or A Boon? We’ll Decide. Forbes. https://www.forbes.com/sites/frederickhess/2023/02/08/will-chatgpt-be-a-blow-to-learning-or-a-boon-well-decide/
  • Hsu, H.H. & Huang, N.F. (2022). Xiao-Shih: a self-enriched question answering bot with machine learning on Chinese-based MOOCs. IEEE Transactions...
  • Ibanez, M., Reyes, L.L.A., Sapinit, R., Hussien, M.A. & Imperial, J.M. (2022). On applicability of neural language models for readability...
  • Jayaraman, J.D. & Black, J. (2022). Effectiveness of an Intelligent Question Answering System for Teaching Financial Literacy: A Pilot...
  • Khot, T., Clark, P., Guerquin, M., Jansen, P. & Sabharwal, A. (2020). QASC: A Dataset for Question Answering via sentence composition....
  • Khushk, A., Zhiying, L., Yi, X. & Zengtian, Z. (2023). Technology Innovation in STEM Education: A Review and Analysis. International Journal...
  • Lee, J., Soleimani, F., Irish, I., Hosmer, J., Soylu, M.Y., Finkelberg, R. & Chatterjee, S. (2022). Predicting cognitive presence in at-scale...
  • Lee, M.C., Chang, J.W. & Chen, J.L. (2014). Detecting ESL/EFL grammatical errors based on n-grams and web resources. Conference name:...
  • Leong, K., Sung, A., Au, D., & Blanchard, C. (2020). A review of the trend of microlearning. Journal of Work-Applied Management, 13(1),...
  • Leydesdorff, L. & Etzkowitz, H. (2003). Conference report: Can ‘the public’ be considered as a fourth helix in university-industry-government...
  • Li, H. (2022). Language models: past, present, and future. Communications of the ACM, 65(7), 56–63. https://doi.org/10.1145/3490443
  • Li, Y., Anastasopoulos, A. and Black, A.W. (2020). Towards Minimal Supervision BERT-Based Grammar Error Correction. Proceedings of the AAAI...
  • Lin, J. (2020). Hybrid translation and language model for micro learning material recommendation. 2020 IEEE 20th International Conference...
  • Liu, Q., Liu, T., Zhao, J., Fang, Q., Ding, W., Wu, Z., Xia, F., et al. (2021). Solving ESL sentence completion questions via pre-trained...
  • López-Belmonte, J., Segura-Robles, A., Cho, W. C., Parra-González, M.E. & Moreno-Guerrero, A. J. (2021). What does literature teach about...
  • Lopez-Ferrero, C., Renau, I., Nazar, R. & Torner, S. (2014). Computer-assisted revision in Spanish academic texts: Peer-assessment. Procedia...
  • Lu, X., Sahay, S., Yu, Z. & Nachman, L. (2021). ACAT-G: An Interactive Learning Framework for Assisted Response Generation. Thirty-Fifth...
  • Makhlouf, J. & Mine, T. (2021). Mining students’ comments to build an automated feedback system. Proceedings of the 13th International...
  • Masala, M., Ruseti, S., Dascalu, M. & Dobre, C. (2021). Extracting and clustering main ideas from student feedback using language models....
  • Meisner, C. (2023). Baylor professors fear students will lose critical thinking skills with ChatGPT. Baylot Lariat. https://baylorlariat.com/2023/02/07/baylor-professors-fear-students-will-lose-critical-thinking-skills-with-chatgpt/
  • Moore, S., Nguyen, H.A., Bier, N., Domadia, T. & Stamper, J. (2022). Assessing the quality of student-generated short answer questions...
  • Murray, B. (2023). ChatGPT forces us to rethink student effort and laziness. Psychology Today.https://www.psychologytoday.com/intl/blog/real-happiness-in-a-digital-world/202301/chatgpt-forces-us-to-rethink-student-effort-and
  • Ndukwe, I.G., Amadi, C.E., Nkomo, L.M. & Daniel, B.K. (2020). Automatic grading system using sentence-BERT network. In I. Bittencourt,...
  • Nehyba, J. & Stefanik, M. (2022). Applications of deep language models for reflective writings. Education and Information Technologies,...
  • Nicula, B., Dascalu, M., Newton, N., Orcutt, E. & McNamara, D.S. (2021). Automated paraphrase quality assessment using recurrent neural...
  • Ondas, S., Hladek, D., Stas, J., Juhar, J., Kovacs, L. & Baksane, E.V. (2015). Semantic roles modeling using statistical language models....
  • Pan, L. (2018). Automatic generation of children’s songs based on machine statistic learning. International Journal of Emerging Technologies...
  • Parasa, N.S., Diwan, C. & Srinivasa, S. (2022). Automatic riddle generation for learning resources. In M., Rodrigo, N., Matsuda, A., Cristea,...
  • Perkmann, M., Tartari, V., McKelvey, M., Autio, E., Broström, A., D’Este, P., Fini, R., et al. (2013). Academic engagement and commercialisation:...
  • Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K. & Zettlemoyer, L. (2018). Deep contextualized word representations....
  • Q.ai (2023). Here Comes the Bing Chatbot - Microsoft’s ChatGPT For Search Has Arrived, Forcing Google’s Hand. Forbes. https://www.forbes.com/sites/qai/2023/02/09/here-comes-the-bing-chatbotmicrosofts-chatgpt-for-search-has-arrived-forcing-googles-hand/?sh=6315ec6110fb
  • Rakovic, M., Sha, L., Nagtzaam, G., Young, N., Stratmann, P., Gasevic, D. & Chen, G. (2022). Towards the automated evaluation of legal...
  • Rosen, P. (2023). ChatGPT’s creator OpenAI has doubled in value since 2021 as the language bot goes viral and Microsoft pours in $10 billion....
  • Rosenfeld, R. (2000). Two decades of statistical language modeling: where do we go from here? Proceedings of the IEEE, 88(8), 1270–1278. https://doi.org/10.1109/5.880083
  • Salim, S. (2023). UAE jobs and ChatGPT: Over 70% workers must learn new skills by 2025, says expert. Khaleej Times. https://www.khaleejtimes.com/jobs/uae-jobs-should-employees-worry-about-chatgpt-other-ai-tools-replacing-them
  • Sanghvi, S. & Westhoff, M. (2022). Education technology: Five trends to watch in the EdTech industry. Mckinsey & Company. https://www.mckinsey.com/industries/education/our-insights/five-trends-to-watch-in-the-edtech-industry
  • Sung, C., Dhamecha, T.I. & Mukhi, N. (2019). Improving short answer grading using transformer-based pre-training. In S. Isotani, E. Millan,...
  • Tang, L., Ke, E., Singh, N., Feng, B., Austin, D., Verma, N. & Drori, I. (2022). Solving probability and statistics problems by probabilistic...
  • Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., et al. (2017). Attention is all you need. arXiv....
  • Wang, Z., Valdez, J., Mallick, D.B. & Baraniuk, R.G. (2022). Towards human-like educational question generation with large language models....
  • Westfall, C. (2023). Educators Battle Plagiarism As 89% Of Students Admit To Using OpenAI’s ChatGPT For Homework. Forbes. https://www.forbes.com/sites/chriswestfall/2023/01/28/educators-battle-plagiarism-as-89-of-students-admit-to-using-open-ais-chatgpt-for-homework/
  • Wise, A.F., Cui, Y. & Jin, W.Q. (2017). Honing in on social learning networks in MOOC forums: examining critical network definition decisions....
  • Wulff, P., Buschhueter, D., Westphal, A., Mientus, L., Nowak, A. & Borowski, A. (2022). Bridging the gap between qualitative and quantitative...
  • Xiao, C., Shi, L., Cristea, A., Li, Z. & Pan, Z. (2022). Fine-grained Main Ideas Extraction and Clustering of Online Course Reviews. In...
  • Xu, S., Ding, W. & Liu, Z. (2020). Automatic Dialogic Instruction Detection for K-12 Online One-on-One Classes. In I. Bittencourt, M....
  • Xu, S., Xu, G., Jia, P., Ding, W., Wu, Z. & Liu, Z. (2021). Automatic Task Requirements Writing Evaluation via Machine Reading Comprehension....
  • Yang, G., Wen, D., Kinshuk, Chen, N.S. & Sutinen, E. (2012). Personalized Text Content Summarizer for Mobile Learning: An Automatic Text...
  • Zhu, X., Wu, H. & Zhang, L. (2022). Automatic short-answer grading via BERT-based deep neural networks. IEEE Transactions on Learning...