Ressources sur les systèmes de génération automatique de textes#
Information
Ressources
Auteurs : Philippe Dessus, Inspé & LaRAC, Univ. Grenoble Alpes, Alexandra Salou, IMSIC, Univ. Aix-Marseille, et Benjamin Serventon, DAPI, Univ. Grenoble Alpes.
Date de création : Juin 2023.
Résumé : Ce Document liste des ressources sur les systèmes de génération de textes (SGAT, ou agents/robots conversationnels), dans la lignée de ChatGPT, dans une perspective éducative.
Voir aussi : Utilisation de systèmes de génération automatique de textes.
Note#
Les abréviations suivantes sont utilisées plus bas : IA : intelligence artificielle ; SGAT : systèmes de génération automatique de textes (IA génératrice de textes) ; LLM : large language models (grands modèles de langage).
Avertissement : La plupart des ressources ci-dessous ne sont pas passées par un processus de relecture par les pairs. Il convient donc de prendre avec précaution les résultats exposés. La nouveauté du thème de recherche et la vivacité du champ rend très difficile la tâche de proposer des ressources fiables.
Sites de ressources générales sur l’IA et les SGAT#
Ressources diverses#
Daily Papers (Hugging Face)
École branchée. Eduquer les élèves à l’IA
N° spécial 136 ERCIM sur les LLM
Liste d’outils#
Arthur, R. (2023). AI Tools for teachers
Applications à base de SGAT#
Type d’outils : https://twitter.com/heykahn/status/1638544259925299203?s=20
“Détecteurs” d’utilisation de SGAT#
Travaux généraux sur l’IA et la société#
Rapports#
Entretiens, opinions#
Barret Bertelloni, M. (2023, 15 juillet). Nick Srniček : « On peut imaginer un agenda radical en matière de nouvelles technologies ». AOC Média. [Accès payant]
Heikkilä, M. (2023, 30 mai). How to talk about AI (even if you don’t know much about AI). MIT Technology Review.
Histoire des SGAT#
Coa et al. (2023, 7 mars). A Comprehensive Survey of AI-Generated Content (AIGC): A History of Generative AI from GAN to ChatGPT. ArXiV.
L’impact environnemental des SGAT#
Anonyme (2023, 11 sept.). Artificial intelligence technology behind ChatGPT was built in Iowa — with a lot of water
Luccioni, A. A., et al. (2022). Estimating the carbon footprint of BLOOM, A 176b parameter language model. ArXiV.
Tomlison, B., et al. (2023, 8 mars). The Carbon Emissions of Writing and Illustrating Are Lower for AI than for Humans. ArXiV.
IA et vie privée#
Solove, D. J. (2024). Artificial Intelligence and Privacy. SSRN Electronic Journal.
Microtravail#
Dzieza, J. (2023, 20 juin). AI is a lot of work. The Verge.
Meaker, M. (2023, 11 sept.). These prisoners are training AI. Wired.
Tubaro, P. & Coville, M. & Le Ludec, C. & Casilli, A. A. (2022). Hidden inequalities: the gendered labour of women on micro-tasking platforms Internet Policy Review, 11(1).
Veselovsky V. et al. (2023, 13 juin)Artificial Artificial Artificial Intelligence: Crowd Workers Widely Use Large Language Models for Text Production Tasks. ArXiV.
Les SGAT comme aide à la recherche#
Dwivedi, Y. K., et al. (2023). Opinion Paper: “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy. International Journal of Information Management, 71. https://doi.org/10.1016/j.ijinfomgt.2023.102642
Grossmann, I., Feinberg, M., Parker, D. C., Christakis, N. A., Tetlock, P. E., & Cunningham, W. A. (2023). AI and the transformation of social science research. Science, 280(6650), 1108–1109. https://doi.org/10.1126/science.adi1778
Hoover, A. (2023, août). Use of AI Is Seeping Into Academic Journals—and It’s Proving Difficult to Detect. Wired.
Liang, W. et al. (2023, oct.). Can large language models provide useful feedback on research papers? A large-scale empirical analysis. ArXiV.
Nam, B. H., & Bai, Q. (2023). ChatGPT and its ethical implications for STEM research and higher education: a media discourse analysis. International Journal of STEM Education, 10(1). https://doi.org/10.1186/s40594-023-00452-5
Srivastava, M. (2023, mars). A day in the life of ChatGPT as an academic reviewer: Investigating the potential of large language model for scientific literature review. ArXiV.
Underwood, T. (2023, mars). Using GPT-4 to measure the passage of time in fiction. Billet de blog
Yao X. et al. (2024, février). Evaluating the efficacy of artificial intelligence tools for the automation of systematic reviews in cancer research: A systematic review. Cancer Epidemiol., 88.
Ziems et al. (2023, avril) Can LLM transform computational social science?.
Travaux généraux sur l’IA et l’éducation#
Rapports#
AMUE (2023, déc.). IA et enseignement supérieur : quels enjeux et impacts ?
Artificial intelligence and teaching, Univ. Monash, Australie
De la Higuera, C., & Iyer, J. (2023). IA pour les enseignants : un manuel ouvert.
DNE-TN2 (2023). Intelligence artificielle et éducation : Apports de la recherche et enjeux pour les politiques publiques. Éducation, numérique et recherche. Paris : MEN.
Office of Educ. Technol. (2023, mai). Artificial intelligence and the future of teaching and learning.
Romero, M. et al. (2023, 13 mars). Enseigner et apprendre à l’ère de l’intelligence artificielle. Canopé).
Unesco (2021). IA et éducation. Guide pour les décideurs politiques
Ressources pédagogiques pratiques sur l’IA#
Goodlad, L. M. E., & Stoerger, S. (2023). Teaching critical AI literacy. Rutgers Univ.
Masure, A. IA et pédagogie : état de l’art
Université de Bordeaux (2023). Intelligences artificielles génératives : ressources
Entretiens critiques sur l’utilisation de l’IA dans l’éducation#
Andronikidis, K. (2023). Meaningful and ethical use of data in schools. Data4Learning Webinar Series. European Schoolnet. Brussels, Belgium [Transcription de webinaires].
Boullier, D. (2023, 7 fév.). Sciences Po a eu raison d’interdire ChatGPT. AOC. Accès payant.
Collin, S. M., & Marceau, E. (2021). L’intelligence artificielle en éducation : enjeux de justice. Formation et Profession, 29(2), 1. https://doi.org/10.18162/fp.2021.a230
Aspects éthiques de l’utilisation de l’IA en éducation#
L’IA génératif (les SGAT) et l’éducation#
Guides généraux d’utilisation des SGAT à l’université#
Atlas, S. (2023). ChatGPT for Higher Education and Professional Development: A Guide to Conversational AI.
Miao, F. (2023). Guidance for generative AI in education and research. Paris: Unesco.
Npuls (2023). Smarter education with AI
Rapports#
Sabzalieva, E. & Valentini, A. (2023). ChatGPT and artificial intelligence in higher education. Paris : UNESCO.
Cours#
Facciolo, L. (2024, Février). You & AI
Stanford (Teaching commons). Defining AI and chatbots
Stanford (Teaching commons). Exploring the pedagogical uses of AI chatbots
Stanford (Teaching commons). Analyzing the implications of AI for your course
Stanford (Teaching commons). Creating your course policy on AI
Revues générales sur les SGAT et l’éducation#
Albadarin, Y. (2023, 9 mai). Usages de chatGPT en éducation.
Amer-Yahia, S., Bonifati, A., Chen, L., Li, G., Shim, K., Xu, J., & Yang, X. (2023). From Large Language Models to Databases and Back: A Discussion on Research and Education. In 28th International Conference on Database Systems for Advanced Applications (DASFAA 2023), Tianjin.
Association EPI (2023, mars). A propos de ChatGPT
Banerjee, P. et al. (2023, Avril). Understanding ChatGPT: Impact Analysis and Path Forward for Teaching Computer Science and Engineering. TecRxiV.
Cuban, L. (2023, 13 juin). ChatGPT is going to change education, not destroy it.
Dawson, P. (2023, 13 juin). Don’t fear the robot. Présentation.
Fawns, T. (2023, 8 sept). Expanding the unit of analysis of learning.
FII Institute (2023). Will ChatGPT give us a lesson in education?. Nature portfolio.
Fraiwan et al. (2023). A Review of ChatGPT Applications in Education, Marketing, Software Engineering, and Healthcare
Futterer, T., Fischer, C., Alekseeva, A., Chen, X., Tate, T., Warschauer, M., & Gerjets, P. (2023). ChatGPT in education: global reactions to AI innovations. Sci Rep, 13(1), 15310. https://doi.org/10.1038/s41598-023-42227-6
Hannigan, T. R., McCarthy, I. P., & Spicer, A. (2023). Beware of botshit: How to manage the epistemic risks of generative chatbots. SSRN Electronic Journal. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4678265
Herft, A. (2023). Guide de l’enseignant. L’usage de ChatGPT
IA et éducation : apports de la recherche et focus sur les IA génératives
Journée de réflexion collective sur l’IA générative. Montréal : UQÀM.
Kahn, S. (2023). The amazing AI super tutor for students and teachers. TED Talk.
Kasneci, E., Seßler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., … Kasneci, G. (2023, January 30). ChatGPT for Good? On Opportunities and Challenges of Large Language Models for Education. https://doi.org/10.35542/osf.io/5er8f
Khosravi, H., Denny, P., Moore, S., & Stamper, J. (2023). Learnersourcing in the age of AI: Student, educator and machine partnerships for content creation. Computers and Education: Artificial Intelligence, 5. https://doi.org/10.1016/j.caeai.2023.100151
Lodge, J. M., Yang, S., Furze, L., & Dawson, P. (2023). It’s not like a calculator, so what is the relationship between learners and generative artificial intelligence? Learning: Research and Practice, 9(2), 117-124. https://doi.org/10.1080/23735082.2023.2261106
Mills, A., Bali, M., & Eaton, L (2023). How do we respond to generative AI in education?. J. Applied Learning & Teaching, 6(1), 1-15.
Mishra, P., Oster, N., & Henriksen, D. (2024). Generative AI, Teacher Knowledge and Educational Research: Bridging Short- and Long-Term Perspectives. TechTrends. https://doi.org/10.1007/s11528-024-00938-1
Mollick, E. R. & Mollick, L. (2022, 13 déc.). New Modes of Learning Enabled by AI Chatbots: Three Methods and Assignments. SSRN.
Morgan, G. (2023, 4 mai). The 5 pathologies of EdTech Discourse about Generative AI
Robinson, G. (2023). Generative AI in Education: An introduction
Rose, R. (2023). ChatGPT in higher education
Van Rooij, I. (2023, 14 janv.). Stop feeding the hype and start resisting
Williamson, B. (2024, Février). AI in education is a public problem. Blog post.
Williamson, B., Molnar, A., & Boninger, F. (2024, Mars). Time for a pause: Without effective public oversight, AI in schools will do more harm than good. Boulder, CO: National Education Policy Center.
Yan, L., Sha, L., Zhao, L., Li, Y., Martinez‐Maldonado, R., Chen, G., Li, X., Jin, Y., & Gašević, D. (2023). Practical and ethical challenges of large language models in education: A systematic scoping review. British Journal of Educational Technology, 55(1), 90-112. https://doi.org/10.1111/bjet.13370
Yang, Z., Wu, J. G., & Xie, H. (2024). Taming Frankenstein’s monster: Ethical considerations relating to generative artificial intelligence in education. Asia Pacific Journal of Education, 1-14. https://doi.org/10.1080/02188791.2023.2300137
SGAT et éducation : questions plus spécifiques#
Impact des SGAT sur les pratiques évaluatives#
Carrefour, Le (2023). 10 stratégies pour éviter le plagiat lors de l’utilisation d’un agent conversationnel (ChatGPT) dans les évaluations. UQUÀM.
Crozat, S. (2023, 26 sept.). IA génératives : la fin des exercices rédactionnels à l’université ?
Francis, N., & Smith, D. (2023). Generative AI in assessment. Guidance staff. Edge Hill Univ.
Hutson, J. & Plate, D. (2023). Working With (Not Against) the Technology: GPT3 and Artificial Intelligence (AI) in College Composition. Lindenwood Univ.
Lang, J. M. (2023, 4 avril)How to Create Compelling Writing Assignments in a ChatGPT Age
Liu, D., & Bridgeman, A. (2023, 8 juin). ChatGPT is old news: How do we assess in the age of AI writing co-pilots?
Lodge, J. M. (2023, 1er mai). Assessment redesign for generative AI: A taxonomy of options and their viability.
Mollick, E. (2023, 1er Juillet). The homework apocalypse. Billet de blog.
Tate, T. P., Steiss, J., Bailey, D., Graham, S., Ritchie, D., Tseng, W., Moon, Y., & Warschauer, M. (2023, 7 déc.). Can AI Provide Useful Holistic Essay Scoring? OSF Preprint. https://doi.org/https://osf.io/preprints/osf/7xpre
TEQSA (2023, sept.). Assessment reform for the age of artificial intelligence
Tricherie et détection d’usage de ChatGPT#
Antoun, W. et al. (2023). Towards a Robust Detection of Language Model-Generated Text: Is ChatGPT that Easy to Detect?. TALN 2023 Conf.
Bogost, I. (2023, 16 mai). The first year of AI College ends in Ruin
Carrefour, Le. 10 stratégies pour éviter le plagiat lors de l’utilisation d’un agent conversationnel dans les évaluations
Chaka, C. (2023). Detecting AI content in responses generated by ChatGPT, YouChat, and Chatsonic: The case of five AI content detection tools. Journal of Applied Learning & Teaching, 6(2).
Eaton, S.E. AI and academic Integrity
Fröhling L, Zubiaga A. (2021). Feature-based detection of automated language models: tackling GPT-2, GPT-3 and Grover. PeerJ Computer Science, 7:e443
Jiang, Z. et al. (2023, 5 mai). Evading Watermark based Detection of AI-Generated Content. ArXiV.
Kan, M. (2023). Instructor Accuses Texas A&M Class of Using ChatGPT, Withholds Grades
Kirchenbauer J. et al. (2023, 9 juin). On the Reliability of Watermarks for Large Language Models.
Klee, M. (2023, 6 juin). She Was Falsely Accused of Cheating With AI — And She Won’t Be the Last
Liang et al. (2023). GPT detectors are biased against non-native English writers
Mindner, L. et al. (2023, Août). Classification of Human- and AI-Generated Texts: Investigating Features for ChatGPT. ArXiV.
Mitrovic, S., Davide, A., & Ayoub, O. (2023). ChatGPT or human? Detect and explain. Explaining decisions of machine learning model for detecting short ChatGPT-generated text. ArXiv preprint.
OpenAI (2023). How can educators respond to students presenting AI-generated content as their own?.
Sadasivan et al. (2023). Can AI-generated text be reliably detected.
Sheinman Orenstrakh et al. (2023, juil.). Detecting LLM-Generated Text in Computing Education: A Comparative Study for ChatGPT Cases. ArXiV.
Šigut, P., & Foltýnek, T. s. (2023). Can We Detect ChatGPT-generated Texts in Czech and Slovak Languages? In A. Horák, P. Rychlý, & A. Rambousek (Eds.), Proc. Recent Advances in Slavonic Natural Language Processing (RASLAN 2023) (pp. 35–43). Tribun EU.
Tang, R. et al. (2023, juin). The Science of Detecting LLM-Generated Texts. ArXiV
Weber-Wulff, D. et al. (2023, 10 Juil.). Testing of Detection Tools for AI-Generated Text. ArXiV.
Yu, P., Chen, J., Feng, X., & Xia, Z. (2023). CHEAT: A Large-scale Dataset for Detecting ChatGPT-writtEn AbsTracts. ArXiv preprint.
Politiques d’usage des SGAT dans les universités#
Artificial intelligence and teaching, Univ. Monash, Australie
Conrad, K. (2023). A blueprint for an AI bill of rights for education
Dunnigan, J., Henriksen, D., Mishra, P., & Lake, R. (2023). “Can we just Please slow it all Down?” School Leaders Take on ChatGPT. TechTrends, 67(6), 878-884. https://doi.org/10.1007/s11528-023-00914-1
Eaton, L. (2023). Classroom policies for AI generative tools
Luo, J. (2024). A critical review of GenAI policies in higher education assessment: a call to reconsider the “originality” of students’ work. Assessment & Evaluation in Higher Education, 1-14. https://doi.org/10.1080/02602938.2024.2309963
Mucharraz y Cano et al. (2023, 1er Fév.). ChatGPT and AI Text Generators: Should Academia Adapt or Resist?. Harvard Business Publishing Education.
Russell Group (2023, 4 Juillet). New principles on use of IA in education.
Stanford Univ. (2023). Generative AI Policy Guidance.
Welker, G. (2023). BBJ: Academic unit at Boston Univ. adopts guidelines for use of generative AI.
Yale Univ. (2023). Resources on ChatGPT and other AI software
Yale Univ. (2023). AI guidance
Les représentations de l’IA par les usagers#
Ada Lovelace Institute, & Alan Turing Institute. (2023). How do people feel about AI? A nationally representative survey of public attitudes to artificial intelligence in Britain. London.
Avry, S., Mondada, F., Liégeois, G., Vaucher, E., Martinez, F., & Dehler Zufferey, J. (2023). ChatGPT. L’avis des enseignant·es. Lausanne: EPFL.
SGAT et pédagogie#
Les manuels génératifs#
Wiley, D. (2023, 6 Juillet). Generative Textbooks-A brief example.
Le tutorat#
Phung et al. (2023, 29 juin). Generative AI for Programming Education: Benchmarking ChatGPT, GPT-4, and Human Tutors. ArXiV.
Les SGAT et le raisonnement scientifique#
Goldstein, J. A., Chao, J., Grossman, S., Stamos, A., & Tomz, M. (2024). How persuasive is AI-generated propaganda? PNAS Nexus, 3(2), pgae034. https://doi.org/10.1093/pnasnexus/pgae034
Nikolic, S., Daniel, S., Haque, R., Belkina, M., Hassan, G. M., Grundy, S., Lyden, S., Neal, P., & Sandison, C. (2023). ChatGPT versus engineering education assessment: a multidisciplinary and multi-institutional benchmarking and analysis of this generative artificial intelligence tool to investigate assessment integrity. European Journal of Engineering Education, 48(4), 559-614. https://doi.org/10.1080/03043797.2023.2213169
Perkowitz, S. (2023, 28 mars). What Does ChatGPT Know About Science?. Nautilus, 49.
Les SGAT et l’enseignement des langues#
Jeon, J., & Lee, S. (2023). Large language models in education: A focus on the complementary relationship between human teachers and ChatGPT. Education and Information Technologies. https://doi.org/10.1007/s10639-023-11834-1
Les SGAT et la formation des enseignants#
Dillig, M., Böhmer, A., Mason, J., Bolaji, S., Isso, I., Stănescu, M. H., Sahin, H., & Kuraner, A. (2024). A Segue From Search to Dialogue. In Transforming Education With Generative AI (pp. 1-34). IGI Press https://doi.org/10.4018/979-8-3693-1351-0.ch002
Les SGAT pour la créativité et la curiosité#
Abdelghani, R. et al. (2022, nov.). GPT-3-driven pedagogical agents for training children’s curious question-asking skills. ArXiV.
Girota, K. et al. (2023, août). Ideas are Dimes a Dozen: Large Language Models for Idea Generation in Innovation. SSRN.
Haase, J. & Hanel, P.H.P. (2023, mars). Artificial muses: Generative Artificial Intelligence Chatbots Have Risen to Human-Level Creativity. ArXiV.
Luchini, S. et al. (2023, oct.)? Automatic Scoring of Creative Problem-Solving with Large Language Models: A Comparison of Originality and Quality Ratings. PsyArXiv.
Mollick, E. (2023, 13 août). Automating creativity. Blog.
Aspects techniques#
Le fonctionnement des SGAT#
Bycroft, B. (2023). LLM Visualization
OpenAI (2023, mars). GPT-4 Technical Report. ArXiV.
Ornes, S. (2023). Modèles massifs de langage. D’où viennent les coups de génie de l’IA ? Pour la Science, 549, 47–51.
Prince, S. J. D. (2023, 8 mai). Understanding deep learning. MIT Press.
Shanahan, M. (2023, fév.). Talking about large language models. ArXiV.
Vig, J. (2019). A Multiscale Visualization of Attention in the Transformer Model. Ann. Meeting of the ACL.
Visual storytelling team & Murgia, M. (2023). Generative AI exists because of the transformer. Financial Times.
Vu, T. et al. (2023, oct.). FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation. ArXiV.
Rétroingénierie des SGAT#
Les corpus d’entraînement des SGAT#
Gao, L. et al. (2020). The Pile: An 800GB Dataset of Diverse Text for Language Modeling. ArXiV.
Les LLM en tant que modèles psychologiques#
Atari, M. et al. (2023, avril). Contextualized Construct Representation: Leveraging Psychometric Scales to Advance Theory-Driven Text Analysis. PsyArXiV.
Atari, M. et al. (2023, sept.). Which humans?. PsyArXiV.
Bajohr, H. (2023). Dumb meaning: ML and artificial semantics. Image., sur le symbol grounding](https://hal.science/hal-04013223v2/file/202303_LivreBlanc_UsagesCreatifsIA_GTnum_Scol_IA_R02.pdf).
Bakhtin, A. et al. (2023). Human-level play in the game of Diplomacy by combining language models with strategic reasoning. Science, 378, 1067-1074.
Blank, I. A. (2023). What are large language models supposed to model? Trends Cogn Sci. https://doi.org/10.1016/j.tics.2023.08.006
Connell, L. & Lynott, D. (2024). What Can Language Models Tell us About Human Cognition? OSF Preprint.
Demszky, et al. (2023). Using large language models in psychology. Nature Reviews Psychology. https://doi.org/10.1038/s44159-023-00241-5
Dillion, D., Tandon, N., Gu, Y., & Gray, K. (2023). Can AI language models replace human participants? Trends Cogn Sci, 27(7), 597-600. https://doi.org/10.1016/j.tics.2023.04.008
Guha, N. et al. (2023). LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models. ArXiV.
Gurnee, W. & Tegmark, M. (2023, 3 oct.). Language Models Represent Space and Time. ArXiV
Ho, N. et al. (2023, 13 juin). Large Language Models Are Reasoning Teachers. ArXiV.
Jones, C., & Bergen, B. (2023, 31 oct.). Does GPT-4 Pass the Turing Test?. ArXiV.
Lucchi, N. (2023). ChatGPT: A Case Study on Copyright Challenges for Generative Artificial Intelligence Systems. European Journal of Risk Regulation, 1-23. doi:10.1017/err.2023.59
Park, P.S. et al. (2023, 28 août). AI Deception: A Survey of Examples, Risks, and Potential Solutions. arXiV.
Rathje, S., Mirea, D.-M., Sucholutsky, I., Marjieh, R., Robertson, C. E., & Bavel, J. J. V. (2024). GPT is an effective tool for multilingual psychological text analysis. PsyArXiv Preprint. https://osf.io/preprints/psyarxiv/sekf5
Saha, D., Brooker, P., Mair, M. and Reeves, S. (2023). Thinking Like a Machine: Alan Turing, Computation and the Praxeological Foundations of AI. Science & Technology Studies. https://doi.org/10.23987/sts.122892
Shi et al. (2023). KokoMind: Can LLMs Understand Social Interactions?
Yax, N. et al. (2023). Studying and improving reasoning in humans and machines. ArXiV.
Zador, A. et al. (2023, 22 mars). Catalysing next-generation Artificial intelligence through NeuroAI. Nature Communications, 14:1597.
Ecrire des prompts#
Akin, F. K. (s.d.). The art of ChatGPT prompting
Eager, B., & Brunton, R. (2023). Prompting Higher Education Towards AI-Augmented Teaching and Learning Practice
Mollick, E. (2023, 26 avril) A guide to prompting AI
Mollick, E. R., & Mollick, L. (2023, 12 juin). Assigning AI: Seven Approaches for Students, with Prompts. SSRN Archive.
Mollick, E. R., & Mollick, L. (2023, 24 mars). Using AI to Implement Effective Teaching Strategies in Classrooms: Five Strategies, Including Prompts. SSRN