Taking the next step with generative artificial intelligence: The transformative role of multimodal large language models in science education
Open accessThe DOI matches its Crossref record; 49 of the 49 references it cites were found in a registry or an index.
Abstract
The integration of Artificial Intelligence (AI), particularly Large Language Model (LLM)-based systems, in education has shown promise in enhancing teaching and learning experiences . However, the advent of Multimodal Large Language Models (MLLMs) like GPT-4 Vision, capable of processing multimodal data including text, sound, and visual inputs, opens a new era of enriched, personalized, and interactive learning landscapes in education. This paper derives a theoretical framework for integrating MLLMs into multimodal learning. This framework serves to explore the transformative role of MLLMs in central aspects of science education by presenting exemplary innovative learning scenarios. Possible applications for MLLMs range from content creation to tailored support for learning, fostering engagement in scientific practices , and providing assessments and feedback. These applications are not limited to text-based and uni-modal formats but can be multimodal, thus increasing personalization, accessibility, and potential learning effectiveness. Despite the many opportunities, challenges such as data protection and ethical considerations become salient, calling for robust frameworks to ensure responsible integration. This paper underscores the necessity for a balanced approach in implementing MLLMs, where the technology complements rather than supplants the educators' roles, ensuring an effective and ethical use of AI in science education. It calls for further research to explore the nuanced implications of MLLMs for educators and to extend the discourse beyond science education to other disciplines. Through developing a theoretical framework for the integration of MLLMs into multimodal learning and exploring the associated potentials, challenges, and future implications, this paper contributes to a preliminary examination of the transformative role of MLLMs in science education and beyond.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
High-Resolution Image Synthesis with Latent Diffusion Models
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01042 · Cited by 20,302
Hattie, J., & Timperley, H. (2007) · Review of Educational Research · doi:10.3102/003465430298487 · Cited by 12,905 · cited 1×
“Assessment is crucial for successful learning trajectories [Hattie 2008; Hattie and Timperley 2007].”
A Comprehensive Survey on Graph Neural Networks
Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.2978386 · Cited by 11,000 · cited 1×
“This kind of reasoning in combination with other types of deep learning and explainability approaches [Borisov et al. 2022a; Rombach et al. 2022; Wu et al. 2020] has the potential to advance science [Wong et al. 2023] and education.”
Kirschner, P. A., Sweller, J., & Clark, R. E. (2006) · Educational Psychologist · doi:10.1207/s15326985ep4102_1 · Cited by 7,776 · cited 4×
“As with assessment, current feedback tools are mostly text-based e.g., Seßler et al. [2023] and, therefore, might not use their full potential to provide effective feedback in science education.”
InfluentialSurvey of Hallucination in Natural Language Generation
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023) · ACM Computing Surveys · doi:10.1145/3571730 · Cited by 5,025 · cited 1×
“Incorporating human feedback through the training of reward models and applying reinforcement learning in the fine-tuning step [Lee et al. 2023b; Ouyang et al. 2022] has significantly enhanced the generation of human-like language that is aligned with social norms and goals and has led to the…”
Survey of Hallucination in Natural Language Generation
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y., Chen, D., Dai, W., et al. (2022) · arXiv · doi:10.48550/arxiv.2202.03629 · Cited by 4,981
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208
Nine Ways to Reduce Cognitive Load in Multimedia Learning
Mayer, R. E., & Moreno, R. (2003) · Educational Psychologist · doi:10.1207/s15326985ep3801_6 · Cited by 4,207
On the Opportunities and Risks of Foundation Models
Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., et al. (2021) · arXiv · doi:10.48550/arxiv.2108.07258 · Cited by 2,676
Llama 2: Open Foundation and Fine-Tuned Chat Models
Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.09288 · Cited by 2,604
PaLM: Scaling Language Modeling with Pathways
Chowdhery, A., Narang, S., Devlin, J., Bosma, M., Mishra, G., Roberts, A., Barham, P., Chung, H. W., Sutton, C., Gehrmann, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.02311 · Cited by 2,590
Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
Zheng, L., Chiang, W.-L., Sheng, Y., Zhuang, S., Wu, Z., Zhuang, Y., Lin, Z., Li, Z., Li, D., Xing, E. P., et al. (2023) · arXiv · doi:10.48550/arxiv.2306.05685 · Cited by 1,888 · cited 1×
“These models have sparked further development, with researchers fine-tuning them and creating open-source variants like Vicuna [Zheng et al. 2023] and Alpaca [Taori et al. 2023].”
Flamingo: a Visual Language Model for Few-Shot Learning
Alayrac, J.-B., Donahue, J., Luc, P., Miech, A., Barr, I., Hasson, Y., Lenc, K., Mensch, A., Millican, K., Reynolds, M., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.14198 · Cited by 1,555
Multimedia learning: Are we asking the right questions?
Mayer, R. E. (1997) · Educational Psychologist · doi:10.1207/s15326985ep3201_1 · Cited by 1,533
Expertise Reversal Effect and Its Implications for Learner-Tailored Instruction
Kalyuga, S. (2007) · Educational Psychology Review · doi:10.1007/s10648-007-9054-3 · Cited by 1,056 · cited 1×
“This would make it possible to avoid expertise reversal effects [Kalyuga 2007].”
Pavlik, J. V. (2023) · Journalism & Mass Communication Educator · doi:10.1177/10776958221149577 · Cited by 972 · cited 1×
“…innovative approaches in areas such as healthcare [Chintagunta et al. 2021; Enarvi et al. 2020], finance [Dowling and Lucey 2023], journalism [Pavlik 2023], creative writing [Yuan et al. 2022], and even in scientific discovery [Romera-Paredes et al. 2023], demonstrating the strength and…”
The Cambridge Handbook of the Learning Sciences
(2014) · doi:10.1017/cbo9781139519526 · Cited by 907
High-Resolution Image Synthesis with Latent Diffusion Models
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2021) · arXiv · doi:10.48550/arxiv.2112.10752 · Cited by 737
Van der Kleij, F. M., Feskens, R. C. W., & Eggen, T. J. H. M. (2015) · Review of Educational Research · doi:10.3102/0034654314564881 · Cited by 714 · cited 2×
“Especially forms of (computer-based) ‘rapid formative assessment’ have shown to be highly effective [Yeh 2010].”
Liu, H., Li, C., Wu, Q., & Lee, Y. J. (2023) · arXiv · doi:10.48550/arxiv.2304.08485 · Cited by 682
PaLM-E: An Embodied Multimodal Language Model
Driess, D., Xia, F., Sajjadi, M. S. M., Lynch, C., Chowdhery, A., Ichter, B., Wahid, A., Tompson, J., Vuong, Q., Yu, T., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.03378 · Cited by 591 · cited 1×
“In summary, combining textual and pictorial information can effectively support learning.”
The Past, Present, and Future of the Cognitive Theory of Multimedia Learning
Mayer, R. E. (2024) · Educational Psychology Review · doi:10.1007/s10648-023-09842-1 · Cited by 576 · cited 1×
“It is crucial for the MLLMs to inherently apply effective multimedia design principles [Mayer 2024], e.g., arising from the signaling principle [van Gog 2014].”
Mathematical discoveries from program search with large language models
Romera-Paredes, B., Barekatain, M., Novikov, A., Balog, M., Kumar, M. P., Dupont, E., Ruiz, F. J. R., Ellenberg, J. S., Wang, P., Fawzi, O., et al. (2024) · Nature · doi:10.1038/s41586-023-06924-6 · Cited by 544
Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
Zhang, Y., Li, Y., Cui, L., Cai, D., Liu, L., Fu, T., Huang, X., Zhao, E., Zhang, Y., Xu, C., et al. (2023) · arXiv · doi:10.48550/arxiv.2309.01219 · Cited by 544 · cited 3×
“However, the capabilities of MLLMs are not limited to images and text; they can also process combinations of text and video data [Li et al. 2023b; Maaz et al. 2023].”
Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
Zhang, Y., Li, Y., Cui, L., Cai, D., Liu, L., Fu, T., Huang, X., Zhao, E., Zhang, Y., Chen, Y., et al. (2025) · Computational Linguistics · doi:10.1162/coli.a.16 · Cited by 534 · cited 3×
“However, the capabilities of MLLMs are not limited to images and text; they can also process combinations of text and video data [Li et al. 2023b; Maaz et al. 2023].”
Promises and challenges of generative artificial intelligence for human learning
Yan, L., Greiff, S., Teuber, Z., & Gašević, D. (2024) · Nature Human Behaviour · doi:10.1038/s41562-024-02004-5 · Cited by 499
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
Shen, Y., Song, K., Tan, X., Li, D., Lu, W., & Zhuang, Y. (2023) · arXiv · doi:10.48550/arxiv.2303.17580 · Cited by 477
HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
Shen, Y., Song, K., Tan, X., Li, D., Lu, W., & Zhuang, Y. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-1657 · Cited by 475
MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models
Zhu, D., Chen, J., Shen, X., Li, X., & Elhoseiny, M. (2023) · arXiv · doi:10.48550/arxiv.2304.10592 · Cited by 469 · cited 1×
“Based on cognitive models of information processing, such as Paivio’s Dual Coding Theory [Paivio 1991], the so-called multimedia effect is the well-supported hypothesis that active processing of both verbal (textual) and non-verbal (visual) information enhances integration into a coherent mental…”
Why Minimally Guided Teaching Techniques Do Not Work: A Reply to Commentaries
SWELLER, J., KIRSCHNER, P. A., & CLARK, R. E. (2007) · Educational Psychologist · doi:10.1080/00461520701263426 · Cited by 434
Wordcraft: Story Writing With Large Language Models
Yuan, A., Coenen, A., Reif, E., & Ippolito, D. (2022) · 27th International Conference on Intelligent User Interfaces · doi:10.1145/3490099.3511105 · Cited by 374 · cited 1×
“…approaches in areas such as healthcare [Chintagunta et al. 2021; Enarvi et al. 2020], finance [Dowling and Lucey 2023], journalism [Pavlik 2023], creative writing [Yuan et al. 2022], and even in scientific discovery [Romera-Paredes et al. 2023], demonstrating the strength and versatility of LLMs.”
Towards hybrid human‐AI learning technologies
Molenaar, I. (2022) · European Journal of Education · doi:10.1111/ejed.12527 · Cited by 348 · cited 1×
“Different modes and degrees of implementation of AI systems are imaginable [Molenaar 2022].”
Out of the laboratory and into the classroom: the future of artificial intelligence in education
Schiff, D. (2021) · AI & SOCIETY · doi:10.1007/s00146-020-01033-8 · Cited by 334
Hornberger, M., Bewersdorff, A., & Nerdel, C. (2023) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2023.100165 · Cited by 276 · cited 1×
“…have to ensure that all learners and educators have access to these potentially powerful tools for learning - they have to provide them with knowledge about AI (AI literacy: Hornberger et al. [2023]; Long and Magerko [2020]) and the competencies they need to successfully employ these AI systems.”
Self-citationConditions that enable effective feedback
Henderson, M., Phillips, M., Ryan, T., Boud, D., Dawson, P., Molloy, E., & Mahoney, P. (2019) · Higher Education Research & Development · doi:10.1080/07294360.2019.1657807 · Cited by 248 · cited 1×
“The design of feedback should always be considered in relation to subsequent related tasks [Henderson et al. 2019].”
Signaling text-picture relations in multimedia learning: A comprehensive meta-analysis
Richter, J., Scheiter, K., & Eitel, A. (2016) · Educational Research Review · doi:10.1016/j.edurev.2015.12.003 · Cited by 242 · cited 1×
“…often have difficulty selecting important information in learning materials, so they need to be supported in their search to avoid unnecessary cognitive load, especially in text-image relations (for a meta-analysis of different studies on the signaling principle, see Richter et al. [2016]).”
Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models
Wu, C., Yin, S., Qi, W., Wang, X., Tang, Z., & Duan, N. (2023) · arXiv · doi:10.48550/arxiv.2303.04671 · Cited by 197 · cited 3×
“Research continues to address these challenges while also improving their explainability to increase trust and reliability in the model outputs [Wu et al. 2023c].”
Challenges and Applications of Large Language Models
Kaddour, J., Harris, J., Mozes, M., Bradley, H., Raileanu, R., & McHardy, R. (2023) · arXiv · doi:10.48550/arxiv.2307.10169 · Cited by 196 · cited 2×
“…approaches in areas such as healthcare [Chintagunta et al. 2021; Enarvi et al. 2020], finance [Dowling and Lucey 2023], journalism [Pavlik 2023], creative writing [Yuan et al. 2022], and even in scientific discovery [Romera-Paredes et al. 2023], demonstrating the strength and versatility of LLMs.”
SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities
Zhang, D., Li, S., Zhang, X., Zhan, J., Wang, P., Zhou, Y., & Qiu, X. (2023) · arXiv · doi:10.48550/arxiv.2305.11000 · Cited by 189
Penedo, G., Malartic, Q., Hesslow, D., Cojocaru, R., Cappelli, A., Alobeidli, H., Pannier, B., Almazrouei, E., & Launay, J. (2023) · arXiv · doi:10.48550/arxiv.2306.01116 · Cited by 164 · cited 1×
“Also, in the open-source landscape, there are strong competitors like Meta’s LLaMA model [Touvron et al. 2023a,b], and the Falcon-40B architecture [Penedo et al. 2023].”
Self-Regulation Behaviors in Underprepared (Developmental) and Regular Admission College Students
Ley, K., & Young, D. B. (1998) · Contemporary Educational Psychology · doi:10.1006/ceps.1997.0956 · Cited by 140 · cited 1×
“This is especially true for students with low self-regulation capabilities [Lan 1998; Ley and Young 1998].”
AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head
Huang, R., Li, M., Yang, D., Shi, J., Chang, X., Ye, Z., Wu, Y., Hong, Z., Huang, J., Liu, J., et al. (2023) · arXiv · doi:10.48550/arxiv.2304.12995 · Cited by 131
GPT-3-Driven Pedagogical Agents to Train Children’s Curious Question-Asking Skills
Abdelghani, R., Wang, Y.-H., Yuan, X., Wang, T., Lucas, P., Sauzéon, H., & Oudeyer, P.-Y. (2024) · International Journal of Artificial Intelligence in Education · doi:10.1007/s40593-023-00340-7 · Cited by 129
RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback
Lee, H., Phatale, S., Mansoor, H., Mesnard, T., Ferret, J., Lu, K., Bishop, C., Hall, E., Carbune, V., Rastogi, A., et al. (2023) · arXiv · doi:10.48550/arxiv.2309.00267 · Cited by 87 · cited 1×
“Incorporating human feedback through the training of reward models and applying reinforcement learning in the fine-tuning step [Lee et al. 2023b; Ouyang et al. 2022] has significantly enhanced the generation of human-like language that is aligned with social norms and goals and has led to the…”
Bewersdorff, A., Seßler, K., Baur, A., Kasneci, E., & Nerdel, C. (2023) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2023.100177 · Cited by 76 · cited 1×
“While LLMs have the potential to significantly impact text-based learning (e.g., Bewersdorff et al. [2023a]; Küchemann et al. [2023]; Seßler et al. [2023]), the advent of MLLMs promises to extend these benefits to include visual, auditory, and other sensory data modalities.”
Self-citationLanguage Models are Realistic Tabular Data Generators
Borisov, V., Seßler, K., Leemann, T., Pawelczyk, M., & Kasneci, G. (2022) · arXiv · doi:10.48550/arxiv.2210.06280 · Cited by 68
Self-citationJung, J., Shin, Y., & Zumbach, J. (2021) · Interactive Learning Environments · doi:10.1080/10494820.2019.1619592 · Cited by 30 · cited 2×
“Therefore, it is crucial to keep learners’ cognitive demands as low as necessary for efficient learning processes to take place [Mayer and Moreno 2003] and further to keep the text and image information understandable for the learner (cf. “pretraining principle”; e.g., Jung et al. [2019]).”
Cited by
Zhai, X. (2025) · Journal of Science Education and Technology · doi:10.1007/s10956-024-10174-0 · Cited by 184
Looking Beyond the Hype: Understanding the Effects of AI on Learning
Bauer, E., Greiff, S., Graesser, A. C., Scheiter, K., & Sailer, M. (2025) · Educational Psychology Review · doi:10.1007/s10648-025-10020-8 · Cited by 153
Zhu, W., Huang, L., Zhou, X., Li, X., Shi, G., Ying, J., & Wang, C. (2025) · International Journal of Human–Computer Interaction · doi:10.1080/10447318.2024.2323277 · Cited by 143
Bai, Y., & Wang, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-08697-6 · Cited by 86
Prompt Engineering in Clinical Practice: Tutorial for Clinicians
Liu, J., Liu, F., Wang, C., & Liu, S. (2025) · Journal of Medical Internet Research · doi:10.2196/72644 · Cited by 76
Metadata
- Authors
- Bewersdorff, A., Hartmann, C., Hornberger, M., Seßler, K., Bannert, M., Kasneci, E., Kasneci, G., Zhai, X., & Nerdel, C.
- Venue
- Learning and Individual Differences
- Publisher
- Elsevier BV
- Published
- 2025-02
- Volume
- 118
- Pages
- 102601
- ISSN
- 1041-6080 (ISSN-L)
- OpenAlex
- W4406232026
- Cited by
- 148
- References
- 49
- FWCI
- 147.49
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.