{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T08:03:23Z","timestamp":1764403403136,"version":"3.41.2"},"reference-count":92,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,9,18]],"date-time":"2024-09-18T00:00:00Z","timestamp":1726617600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Large language models have been shown to excel in many different tasks across disciplines and research sites. They provide novel opportunities to enhance educational research and instruction in different ways such as assessment. However, these methods have also been shown to have fundamental limitations. These relate, among others, to hallucinating knowledge, explainability of model decisions, and resource expenditure. As such, more conventional machine learning algorithms might be more convenient for specific research problems because they allow researchers more control over their research. Yet, the circumstances in which either conventional machine learning or large language models are preferable choices are not well understood. This study seeks to answer the question to what extent either conventional machine learning algorithms or a recently advanced large language model performs better in assessing students' concept use in a physics problem-solving task. We found that conventional machine learning algorithms in combination outperformed the large language model. Model decisions were then analyzed via closer examination of the models' classifications. We conclude that in specific contexts, conventional machine learning can supplement large language models, especially when labeled data is available.<\/jats:p>","DOI":"10.3389\/frai.2024.1408817","type":"journal-article","created":{"date-parts":[[2024,9,18]],"date-time":"2024-09-18T05:10:28Z","timestamp":1726636228000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["David vs. Goliath: comparing conventional machine learning and a large language model for assessing students' concept use in a physics problem"],"prefix":"10.3389","volume":"7","author":[{"given":"Fabian","family":"Kieser","sequence":"first","affiliation":[]},{"given":"Paul","family":"Tschisgale","sequence":"additional","affiliation":[]},{"given":"Sophia","family":"Rauh","sequence":"additional","affiliation":[]},{"given":"Xiaoyu","family":"Bai","sequence":"additional","affiliation":[]},{"given":"Holger","family":"Maus","sequence":"additional","affiliation":[]},{"given":"Stefan","family":"Petersen","sequence":"additional","affiliation":[]},{"given":"Manfred","family":"Stede","sequence":"additional","affiliation":[]},{"given":"Knut","family":"Neumann","sequence":"additional","affiliation":[]},{"given":"Peter","family":"Wulff","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2024,9,18]]},"reference":[{"key":"B1","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint arXiv:2303.08774"},{"key":"B2","article-title":"Transformer models: an introduction and catalog","author":"Amatriain","year":"2023","journal-title":"arXiv preprint arXiv:2302.07730"},{"journal-title":"Bridging the stem skills gap: employer\/educator collaboration in New York","year":"2017","author":"Armour-Garb","key":"B3"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1145\/3442188.3445922","article-title":"\u201cOn the dangers of stochastic parrots,\u201d","author":"Bender","year":"2021","journal-title":"FAccT"},{"key":"B5","doi-asserted-by":"publisher","first-page":"100081","DOI":"10.1016\/j.caeai.2022.100081","article-title":"Machine learning based feedback on textual student answers in large courses","volume":"3","author":"Bernius","year":"2022","journal-title":"Comput. Educ"},{"key":"B6","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1007\/s11133-014-9277-9","article-title":"Humanist interpretation versus coding text samples","volume":"37","author":"Biernacki","year":"2014","journal-title":"Qual. Sociol"},{"journal-title":"Pattern Recognition and Machine Learning","year":"2006","author":"Bishop","key":"B7"},{"key":"B8","doi-asserted-by":"publisher","first-page":"ep430","DOI":"10.30935\/cedtech\/13176","article-title":"Chatgpt in physics education: a pilot study on easy-to-implement activities","volume":"15","author":"Bitzenbauer","year":"2023","journal-title":"Contempor. Educ. Technol"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.69048550\/arXiv.2304.00612","article-title":"Eight things to know about large language models","author":"Bowman","year":"2023","journal-title":"arXiv preprint"},{"key":"B10","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"arXiv preprint arXiv:2005.14165"},{"key":"B11","article-title":"Sparks of artificial general intelligence: early experiments with gpt-4","author":"Bubeck","year":"2023","journal-title":"arXiv preprint arXiv:2303.12712"},{"key":"B12","doi-asserted-by":"publisher","first-page":"183","DOI":"10.1126\/science.aal4230","article-title":"Semantics derived automatically from language corpora contain human-like biases","volume":"356","author":"Caliskan","year":"2017","journal-title":"Science"},{"key":"B13","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1613\/jair.953","article-title":"Smote: synthetic minority over-sampling technique","volume":"16","author":"Chawla","year":"2002","journal-title":"J. Artif. Intell. Res"},{"key":"B14","article-title":"Xplainllm: A qa explanation dataset for understanding llm decision-making","author":"Chen","year":"2023","journal-title":"arXiv preprint arXiv:2311.08614"},{"key":"B15","doi-asserted-by":"publisher","first-page":"37","DOI":"10.18608\/jla.2023.7801","article-title":"The effects of explanations in automated essay scoring systems on student trust and motivation","volume":"10","author":"Conijn","year":"2023","journal-title":"J. Lear. Anal"},{"key":"B16","doi-asserted-by":"publisher","first-page":"2191","DOI":"10.1016\/j.joule.2023.09.004","article-title":"The growing energy footprint of artificial intelligence","volume":"7","author":"de Vries","year":"2023","journal-title":"Joule"},{"key":"B17","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv preprint arXiv:1810.04805"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-45014-9_1","author":"Dietterich","year":"2000","journal-title":"Ensemble Methods in Machine Learning"},{"key":"B19","doi-asserted-by":"publisher","first-page":"020106","DOI":"10.1103\/PhysRevSTPER.11.020106","article-title":"Conceptual problem solving in high school physics","volume":"11","author":"Docktor","year":"2015","journal-title":"Phys. Rev. Special Topics"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1145\/3531146.3533234","article-title":"\u201cMeasuring the carbon intensity of ai in cloud instances,\u201d","author":"Dodge","year":"2022","journal-title":"Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency"},{"key":"B21","article-title":"Enhancing physics learning with chatgpt, bing chat, and bard as agents-to-think-with: a comparative case study","author":"dos Santos","year":"2023","journal-title":"arXiv preprint arXiv:2306.00724"},{"key":"B22","article-title":"Do multilingual language models think better in English?","author":"Etxaniz","year":"2023","journal-title":"arXiv preprint arXiv:2308.01223"},{"key":"B23","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1002\/sce.20295","article-title":"The importance of learning to make assumptions","volume":"93","author":"Fortus","year":"2008","journal-title":"Sci. Educ"},{"journal-title":"Wissen und Probleml\u00f6sen: Eine empirische Untersuchung des wissenszentrierten Probleml\u00f6sens im Gebiet der Elektizit\u00e4tslehre auf der Grundlage des Experten-Novizen-Vergleichs [Knowledge and problem solving: An empirical investigation of knowledge-centred problem solving in the field of electricity based on expert-novice comparison","year":"2001","author":"Friege","key":"B24"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1089","DOI":"10.1080\/09500690600930972","article-title":"Exploring the development of conceptual understanding through structured problem-solving in physics","volume":"29","author":"Gaigher","year":"2007","journal-title":"Int. J. Sci. Educ"},{"key":"B26","article-title":"Teaching vs. learning: changing perspectives on problem solving in physics instruction","author":"Gerace","year":"2005","journal-title":"arXiv preprint physics\/0508131"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40594-018-0110-y","article-title":"Electronixtutor: an intelligent tutoring system with multiple learning resources for electronics","volume":"5","author":"Graesser","year":"2018","journal-title":"Int. J. STEM Educ"},{"key":"B28","doi-asserted-by":"publisher","first-page":"035021","DOI":"10.1088\/1361-6552\/acc299","article-title":"Chatgpt and the frustrated socrates","volume":"58","author":"Gregorcic","year":"2023","journal-title":"Phys. Educ"},{"key":"B29","doi-asserted-by":"publisher","first-page":"100170","DOI":"10.1016\/j.caeai.2023.100170","article-title":"Exploring the potential of ai-chatbots in organic chemistry: an assessment of chatgpt and bard","volume":"5","author":"Hallal","year":"2023","journal-title":"Comput. Educ"},{"key":"B30","first-page":"411","article-title":"spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing","volume":"7","author":"Honnibal","year":"2017","journal-title":"To Appear"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1147","DOI":"10.1119\/1.1763175","article-title":"Resource letter rps-1: research in problem solving","volume":"72","author":"Hsu","year":"2004","journal-title":"Am. J. Phys"},{"key":"B32","article-title":"Look before you leap: an exploratory study of uncertainty measurement for large language models","author":"Huang","year":"2023","journal-title":"arXiv preprint arXiv:2307.10236"},{"key":"B33","doi-asserted-by":"publisher","first-page":"551","DOI":"10.1002\/(SICI)1098-2736(199708)34:6<551::AID-TEA2>3.0.CO;2-M","article-title":"Effect of explicit problem solving instruction on high school students' problem-solving performance and conceptual understanding of physics","volume":"34","author":"Huffman","year":"1997","journal-title":"J. Res. Sci. Teach"},{"volume-title":"Cut the bull... detecting hallucinations in large language models","year":"2023","author":"Hughes","key":"B34"},{"key":"B35","doi-asserted-by":"publisher","first-page":"284","DOI":"10.1007\/s10956-015-9593-1","article-title":"Identifying 21st century stem competencies using workplace data","volume":"25","author":"Jang","year":"2015","journal-title":"J. Sci. Educ. Technol"},{"key":"B36","article-title":"\u201cSpeech and language processing,\u201d","volume-title":"Always Learning","author":"Jurafsky","year":"2014"},{"key":"B37","doi-asserted-by":"publisher","first-page":"102274","DOI":"10.1016\/j.lindif.2023.102274","article-title":"Chatgpt for good? On opportunities and challenges of large language models for education","volume":"103","author":"Kasneci","year":"2023","journal-title":"Learn. Individ. Differ"},{"key":"B38","doi-asserted-by":"publisher","first-page":"020150","DOI":"10.1103\/PhysRevPhysEducRes.19.020150","article-title":"Educational data augmentation in physics education research using chatgpt","volume":"19","author":"Kieser","year":"2023","journal-title":"Phys. Rev. Phys. Educ. Res"},{"key":"B39","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1119\/1.1484151","article-title":"Students do not overcome conceptual difficulties after solving 1000 traditional problems","volume":"70","author":"Kim","year":"2002","journal-title":"Am. J. Phys"},{"key":"B40","article-title":"\u201cLarge language models are zero-shot reasoners,\u201d","author":"Kojima","year":"2022","journal-title":"36th Conference on Neural Information Processing Systems (neurips 2022)"},{"key":"B41","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1103\/PhysRevPhysEducRes.19.010132","article-title":"Could an artificial-intelligence agent pass an introductory physics course?","volume":"19","author":"Kortemeyer","year":"2023","journal-title":"Phys. Rev. Phys. Educ. Res"},{"key":"B42","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-642-37827-0","volume-title":"Methoden in der naturwissenschaftsdidaktischen Forschung","author":"Kr\u00fcger","year":"2014"},{"key":"B43","first-page":"199","article-title":"\u201cUnreflected acceptance-investigating the negative consequences of chatgpt-assisted problem solving in physics education,\u201d","volume-title":"HHAI 2024: Hybrid Human AI Systems for the Social Good","author":"Krupp","year":"2023"},{"key":"B44","doi-asserted-by":"publisher","first-page":"020128","DOI":"10.1103\/PhysRevPhysEducRes.19.020128","article-title":"Can chatgpt support prospective teachers in physics task development?","volume":"19","author":"K\u00fcchemann","year":"2023","journal-title":"Phys. Rev. Phys. Educ. Res"},{"key":"B45","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/s41586-023-06668-3","article-title":"Human-like systematic generalization through a meta-learning neural network","volume":"623","author":"Lake","year":"2023","journal-title":"Nature"},{"key":"B46","doi-asserted-by":"publisher","first-page":"159","DOI":"10.2307\/2529310","article-title":"The measurement of observer agreement for categorical data","volume":"33","author":"Landis","year":"1977","journal-title":"Biometrics"},{"key":"B47","first-page":"1","article-title":"Imbalanced-learn: a python toolbox to tackle the curse of imbalanced datasets in machine learning","volume":"18","author":"Lema\u00eetre","year":"2017","journal-title":"J. Mach. Lear. Res"},{"key":"B48","doi-asserted-by":"publisher","first-page":"1495","DOI":"10.1119\/1.18409","article-title":"Using qualitative problem-solving strategies to highlight the role of conceptual knowledge in solving problems","volume":"64","author":"Leonard","year":"1996","journal-title":"Am. J. Phys"},{"key":"B49","doi-asserted-by":"publisher","first-page":"100140","DOI":"10.1016\/j.caeai.2023.100140","article-title":"Can large language models write reflectively","volume":"4","author":"Li","year":"2023","journal-title":"Comput. Educ"},{"key":"B50","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1145\/3236386.3241340","article-title":"The mythos of model interpretability: In machine learning, the concept of interpretability is both important and slippery","volume":"16","author":"Lipton","year":"2018","journal-title":"Queue"},{"key":"B51","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3560815","article-title":"Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing","volume":"55","author":"Liu","year":"2021","journal-title":"ACM Comput. Surv"},{"key":"B52","article-title":"Explainable ai for trees: From local explanations to global understanding","author":"Lundberg","year":"2019","journal-title":"arXiv preprint arXiv:1905.04610"},{"key":"B53","doi-asserted-by":"publisher","first-page":"127","DOI":"10.1162\/daed_a_01905","article-title":"Human language understanding &reasoning","volume":"151","author":"Manning","year":"2022","journal-title":"Daedalus"},{"key":"B54","doi-asserted-by":"publisher","first-page":"020124","DOI":"10.1103\/PhysRevSTPER.6.020124","article-title":"Surveying graduate students' attitudes and approaches to problem solving","volume":"6","author":"Mason","year":"2010","journal-title":"Phys. Rev. Special Topics"},{"key":"B55","doi-asserted-by":"publisher","first-page":"437","DOI":"10.1162\/COLI_a_00227","article-title":"The unified and holistic method gamma (gamma) for inter-annotator agreement measure and alignment","volume":"41","author":"Mathet","year":"2015","journal-title":"Comput. Lingu"},{"key":"B56","article-title":"\u201cQualitative content analysis,\u201d","author":"Mayring","year":"2000","journal-title":"Forum: Qualitative Social Research"},{"key":"B57","article-title":"Efficient estimation of word representations in vector space","author":"Mikolov","year":"2013","journal-title":"arXiv preprint arXiv:1301.3781"},{"key":"B58","article-title":"Comparing humans, gpt-4, and gpt-4v on abstraction and reasoning tasks","author":"Mitchell","year":"2023","journal-title":"arXiv preprint arXiv:2311.09247"},{"key":"B59","doi-asserted-by":"publisher","first-page":"429","DOI":"10.1007\/s10972-012-9291-2","article-title":"Views about learning physics held by physics teachers with differing approaches to teaching physics","volume":"23","author":"Mulhall","year":"2012","journal-title":"J. Sci. Teacher Educ"},{"journal-title":"Physics doctorates: skills used and satisfaction with employment","year":"2020","author":"Mulvey","key":"B60"},{"key":"B61","doi-asserted-by":"publisher","first-page":"323","DOI":"10.1080\/00107510500052444","article-title":"Power laws, pareto distributions and zipf's law","volume":"46","author":"Newman","year":"2005","journal-title":"Contemp. Phys"},{"key":"B62","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Lear. Res"},{"key":"B63","doi-asserted-by":"publisher","first-page":"034005","DOI":"10.1088\/1361-6404\/aa538f","article-title":"The german physics olympiad\u2014identifying and inspiring talents","volume":"38","author":"Petersen","year":"2017","journal-title":"Eur. J. Phys"},{"key":"B64","doi-asserted-by":"publisher","first-page":"025701","DOI":"10.1088\/1361-6404\/ad1420","article-title":"How understanding large language models can inform the use of chatgpt in physics education","volume":"45","author":"Polverini","year":"2024","journal-title":"Eur. J. Phys"},{"key":"B65","doi-asserted-by":"publisher","DOI":"10.1515\/9781400828678","author":"Polya","year":"1945","journal-title":"How to Solve It"},{"key":"B66","doi-asserted-by":"publisher","first-page":"25","DOI":"10.5120\/ijca2018917395","article-title":"Text mining: use of tf-idf to examine the relevance of words to documents","volume":"181","author":"Qaiser","year":"2018","journal-title":"Int. J. Comput. Applic"},{"key":"B67","doi-asserted-by":"publisher","first-page":"638","DOI":"10.21105\/joss.00638","article-title":"Mlxtend: Providing machine learning and data science utilities and extensions to python's scientific computing stack","volume":"3","author":"Raschka","year":"2018","journal-title":"J. Open Source Softw"},{"key":"B68","doi-asserted-by":"publisher","DOI":"10.1201\/9781003206743","author":"Rauf","year":"2021","journal-title":"Physics of Data Science and Machine Learning"},{"journal-title":"The importance of an enhanced problem representation: on the role of elaborations in physics problem solving","year":"1997","author":"Savelsbergh","key":"B69"},{"key":"B70","article-title":"Best of both worlds: local and global explanations with human-understandable concepts","author":"Schrouff","year":"2021","journal-title":"arXiv preprint arXiv:2106.08641"},{"key":"B71","doi-asserted-by":"publisher","first-page":"e35237","DOI":"10.7759\/cureus.35237","article-title":"Applicability of chatgpt in assisting to solve higher order problems in pathology","volume":"15","author":"Sinha","year":"2023","journal-title":"Cureus"},{"key":"B72","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-02169-5","volume-title":"Argumentation Mining","author":"Stede","year":"2019"},{"key":"B73","doi-asserted-by":"publisher","first-page":"17","DOI":"10.55529\/ijitc.31.17.22","article-title":"Use chat gpt to solve programming bugs","volume":"3","author":"Surameery","year":"2023","journal-title":"Int. J. Inf. Technol. Comput. Eng"},{"key":"B74","doi-asserted-by":"publisher","first-page":"020123","DOI":"10.1103\/PhysRevPhysEducRes.19.020123","article-title":"Integrating artificial intelligence-based methods into qualitative research in physics education research: a case for computational grounded theory","volume":"19","author":"Tschisgale","year":"","journal-title":"Phys. Rev. Phys. Educ. Res"},{"key":"B75","doi-asserted-by":"publisher","DOI":"10.17605\/OSF.IO\/D68CH","article-title":"Computational grounded theory in physics education research","author":"Tschisgale","year":"","journal-title":"Open Sci. Framework"},{"key":"B76","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1186\/s43031-024-00102-y","article-title":"Are science competitions meeting their intentions? A case study on affective and cognitive predictors of success in the physics olympiad","volume":"6","author":"Tschisgale","year":"2024","journal-title":"Discipl. Interdisc. Sci. Educ. Res"},{"key":"B77","doi-asserted-by":"publisher","first-page":"07356331231191174","DOI":"10.1177\/07356331231191174","article-title":"Who's the best detective? Large language models vs. traditional machine learning in detecting incoherent fourth grade math answers","volume":"61","author":"Urrutia","year":"2023","journal-title":"J. Educ. Comput. Res"},{"key":"B78","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1080\/00461520.2011.611369","article-title":"The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems","volume":"46","author":"VanLehn","year":"2011","journal-title":"Educ. Psychol"},{"key":"B79","first-page":"6000","article-title":"\u201cAttention is all you need: Conference on neural information processing systems,\u201d","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B80","article-title":"A multiscale visualization of attention in the transformer model","author":"Vig","year":"2019","journal-title":"ArXiv, abs\/1906.05714"},{"key":"B81","doi-asserted-by":"publisher","first-page":"010152","DOI":"10.1103\/PhysRevPhysEducRes.20.010152","article-title":"Exploring generative AI assisted feedback writing for students' written responses to a physics conceptual question with prompt engineering and few-shot learning","volume":"20","author":"Wan","year":"2024","journal-title":"Phys. Rev. Phys. Educ. Res"},{"key":"B82","article-title":"Finetuned language models are zero-shot learners","author":"Wei","year":"2021","journal-title":"ArXiv, abs\/2109.01652"},{"key":"B83","article-title":"Emergent abilities of large language models","author":"Wei","year":"2022","journal-title":"arXiv preprint arXiv:2206.07682"},{"key":"B84","article-title":"Larger language models do in-context learning differently","author":"Wei","year":"2023","journal-title":"arXiv preprint arXiv:2303.03846"},{"year":"2023","author":"West","key":"B85"},{"key":"B86","article-title":"A prompt pattern catalog to enhance prompt engineering with chatgpt","author":"White","year":"2023","journal-title":"arXiv preprint arXiv:2302.11382"},{"key":"B87","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-demos.6","article-title":"\u201cTransformers: state-of-the-art natural language processing,\u201d","author":"Wolf","year":"2020","journal-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations"},{"key":"B88","doi-asserted-by":"publisher","first-page":"14325","DOI":"10.1007\/s10639-022-11531-5","article-title":"Network analysis of terms in the natural sciences insights from wikipedia through natural language processing and network analysis","volume":"28","author":"Wulff","year":"2023","journal-title":"Educ. Inf. Technol"},{"key":"B89","doi-asserted-by":"publisher","first-page":"035027","DOI":"10.1088\/1361-6552\/acc5cf","article-title":"The death of the short-form physics essay in the coming ai revolution","volume":"58","author":"Yeadon","year":"2023","journal-title":"Phys. Educ"},{"key":"B90","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1080\/09500690500336973","article-title":"Current realities and future possibilities: Language and science literacy\u2013empowering research and informing instruction","volume":"28","author":"Yore","year":"2006","journal-title":"Int. J. Sci. Educ"},{"key":"B91","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1007\/s13042-010-0001-0","article-title":"Understanding bag-of-words model: a statistical framework","volume":"1","author":"Zhang","year":"2010","journal-title":"Int. J. Mach. Learn. Cyber"},{"key":"B92","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3639372","article-title":"Explainability for large language models: a survey","volume":"15","author":"Zhao","year":"2023","journal-title":"ACM Trans. Intell. Syst. Technol"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2024.1408817\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,18]],"date-time":"2024-09-18T05:10:36Z","timestamp":1726636236000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2024.1408817\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,18]]},"references-count":92,"alternative-id":["10.3389\/frai.2024.1408817"],"URL":"https:\/\/doi.org\/10.3389\/frai.2024.1408817","relation":{},"ISSN":["2624-8212"],"issn-type":[{"type":"electronic","value":"2624-8212"}],"subject":[],"published":{"date-parts":[[2024,9,18]]},"article-number":"1408817"}}