Pith. sign in

REVIEW 4 major objections 6 minor 64 references

Education in the Era of Neurosymbolic AI

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read Combining neural LLMs with structured knowledge graphs inside embodied pedagogical agents can deliver fine-grained, scalable personalized tutoring, the paper argues.

desk verdict A well-grounded vision paper that overreaches: the NaPA architecture is a useful synthesis, but the central personalization claim rests on an untested diagnostic assumption, and Section 4's explorations are anecdotal rather than evidence. read the letter →

arxiv 2411.12763 v1 pith:G6I4EEGY submitted 2024-11-16 cs.HC cs.AIcs.CY

classification cs.HCcs.AIcs.CY
keywords neurosymbolicAIpedagogicalagentsknowledgegraphslargelanguagemodelspersonalizedlearningretrieval-augmentedgenerationeducationaltechnologyadaptive
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that education is about to be reshaped by neurosymbolic AI, and proposes a concrete system: a pedagogical agent that combines large language models with structured knowledge graphs to deliver personalized instruction. The hybrid, called a NaPA, is claimed to diagnose each learner's current understanding at a fine-grained level, identify missing prerequisite knowledge, and then choose the right pedagogical move—retrieval practice, spaced repetition, a different modality, or a native-language explanation—so that the whole is greater than its parts. The evidence so far is exploratory: the authors show that LLMs with retrieval-augmented generation over a curriculum knowledge graph can reorganize curricula and adapt language to different learner personas, but they do not yet measure learning outcomes. If the central claim holds, deeply adaptive tutoring could become scalable and widely accessible.

What carries the argument

The load-bearing object is the NaPA architecture, a hybrid in which a symbolic layer (educational and personal knowledge graphs) supplies structured domain knowledge and learner state, a neural layer (LLMs with retrieval-augmented generation) supplies fluent generation and multimodal translation, and an embodied pedagogical agent supplies social presence and instructional interaction. RAG is the mechanism that ties the graph to the generator: relevant facts are retrieved from the KG and folded into the LLM prompt, so answers stay grounded and less prone to hallucination. The pedagogical frameworks—retrieval practice, the testing effect, spaced repetition, and interleaving—are the decision rules that tell the agent which instructional move to make once a gap is identified.

What would settle it

Run a controlled study in which the NaPA diagnoses learners' knowledge gaps from their interaction data and the same learners take an independent concept-inventory test; if the system's gap predictions do not match the inventory at a useful level, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim, stated plainly, is that a neurosymbolic AI-augmented pedagogical agent will be able to interpret complex human concepts and contexts, employ advanced problem-solving strategies grounded in established pedagogical frameworks, and understand when to use each method to produce a sum greater than the constituent components. Concretely, the NaPA is meant to couple the conversational fluency of an LLM with the structured ground truth of an educational knowledge graph and a learner's personal knowledge graph, so it can detect knowledge gaps, adapt curriculum and pacing, switch modalities on demand, and translate content across languages. The paper also claims the embodied agent is not decoration: the social presence of a pedagogical agent activates learning processes described by CASTLE theory, and evidence-based techniques such as retrieval practice and spaced repetition give the system its instructional teeth. The reported explorations—zero-shot generation, persona-based prompting, RAG over a curated curriculum KG, and curriculum redesign with and without pedagogical prompts—support the feasibility of the components, while the full integrated behavior remains a proposal.

Load-bearing premise

The whole loop depends on the system being able to tell, from a learner's interactions, what that learner actually knows and where the gaps are; if that diagnosis is unreliable, every personalized intervention built on it is unreliable too.

Editorial extensions

If this is right

  • A learner who misses a prerequisite concept could be offered a targeted review before new material is introduced, reordering the curriculum in real time.
  • The same content could be rendered as text, audio, diagrams, or captions, making lectures and notes accessible to learners with visual or hearing impairments.
  • Deployable on a device with an internet connection, such agents could bring personalized instruction to regions without enough human tutors or specialist teachers.
  • Educators could offload routine content generation and assessment to the agent and spend more time on deep engagement with students.
  • If a curriculum knowledge graph exists for a domain, the system should be adaptable to that domain without retraining the underlying model.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A decisive test the paper does not report: compare the NaPA's fine-grained gap diagnosis against an independent concept inventory; the personalization loop only works as well as that diagnosis.
  • The architecture implies that knowledge graphs must be actively curated and updated; stale or incomplete graphs would reintroduce the factual errors the symbolic layer is meant to correct.
  • The same graph-grounded generation pipeline could be repurposed for automated assessment and feedback, not just content delivery, which would extend the proposal to a wider class of educational tasks.
  • Because the system selects pedagogy based on learner state, it could serve as a testbed for which instructional strategies work for which learners, generating evidence about personalization itself.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper argues that combining large language models (LLMs) with knowledge graphs (KGs) and embodied pedagogical agents, which it calls NaPAs (Neurosymbolic AI-augmented Pedagogical Agents), will enable deeply personalized, adaptive, and multimodal education at scale. It reviews the foundations of LLMs, KGs, pedagogical agents, and evidence-based pedagogy (retrieval practice, spacing, interleaving), then sketches a hybrid architecture. The manuscript reports 'preliminary explorations' using zero-shot, persona-based, and RAG-augmented prompting to generate educational content and reorganize curricula, and concludes with broad claims about transformative impact on accessibility, equity, and learning outcomes. The paper's stated aim is to discuss the rationale for the system design and preliminary findings, but the abstract and conclusion present future capabilities as near-certainties.

Significance. If the central claims were validated, this work could contribute meaningfully to AIED by connecting KGs, LLMs, and pedagogical agents within a coherent vision. The paper usefully synthesizes several literatures, including CASTLE theory and meta-analytic evidence for retrieval practice, and it makes a credible case that NAI is a promising direction for personalized learning. The authors are also transparent in labeling their studies as exploratory and linking to a public repository. However, the evidence presented is far too weak to support the paper's strong assertions about diagnostic accuracy, adaptive decision-making, and social impact. The value of the paper is currently as a position/vision statement, not as an empirically grounded system proposal.

major comments (4)
  1. [Section 4] The exploratory studies do not evaluate the proposed NaPA system or its central diagnostic capability. The experiments involve standard LLM prompting (zero-shot, persona-based, and RAG) to generate educational modules and reorganize curricula, with no learner participants, no pre/post measures of learning, no comparison against baseline systems, and no quantitative metrics. Consequently, the claims of 'significant improvement' and demonstrated 'ability to organize' curricula are unsupported. This is load-bearing because the paper's central claims in Sections 3.1, 3.2, and 5 depend on the system accurately diagnosing learner knowledge states and selecting adaptive instructional actions.
  2. [Section 3.1] The claim that educational KGs 'can enable accurate detection of the learner's current knowledge state' is asserted without specifying an inference algorithm, a source of ground truth, or any validation. Since the entire personalization loop is closed by this diagnostic step, the paper must either provide evidence for this capability or explicitly reframe it as an open research question. As written, the assertion is an untested premise rather than a supported finding.
  3. [Section 4 and Figure 1] The paper conflates LLM-generated content with NAI system behavior. The RAG experiments use a manually curated KG to ground generation, but this is not the proposed NaPA architecture, which also includes student models, personal KGs, and multimodal translation. The paper does not describe how these components interact, how the student model is updated, or how the agent decides among pedagogical strategies. Without a concrete architecture and a description of the decision policy, the conclusion's claim that NaPAs 'will understand when to use the various methods' (Section 5) is untestable and currently unsupported.
  4. [Section 5 and Section 1] The social-impact claims (e.g., addressing SDG 4, accessibility for underprivileged populations) are speculative and unexamined. The footnote in Section 3.2 acknowledges a 'significant hurdle' but does not analyze feasibility, cost, infrastructure requirements, or risks of bias and equity. These claims are presented as inevitable outcomes, which overstates what can be concluded from the reported preliminary explorations.
minor comments (6)
  1. [Section 1] The text contains 'Figure!1' which should read 'Figure 1'.
  2. [Section 3.2, Footnote 1] The footnote is incomplete: the sentence ends with 'Project Connect Unicef.' without completing the thought or connecting the initiatives to the preceding point about the significant hurdle.
  3. [Section 3.2] The phrase 'for example sourced from the personal KG' is awkwardly placed; consider rephrasing for clarity.
  4. [Section 5] The conclusion repeats earlier claims almost verbatim (e.g., 'delivering multimodal, multilingual, inclusive content') without adding synthesis or actionable next steps; a more focused closing would strengthen the paper.
  5. [Section 4] The paper states 'we observed significant improvement in the quality and relevance' without defining 'quality' or 'relevance' or providing any operationalization; please specify the evaluation criteria or remove 'significant'.
  6. [General] The paper is unclear about its genre: position paper, system proposal, or empirical study. The authors should explicitly state the intended contribution and scope, especially given the mismatch between the strong abstract and the exploratory evidence.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the paper is a forward-looking position piece whose claims are not reduced by construction to its own inputs.

full rationale

This is a position/vision paper rather than a derivation chain. The central claims are explicitly forward-looking ('NAI-powered education systems will be capable', 'NaPAs will understand when to use the various methods') and are not derived from equations, fitted parameters, or a formal model. The exploratory studies in Section 4 use zero-shot, persona-based, and RAG-augmented LLM prompting to generate educational content and reorganize curricula, and the paper reports qualitative improvements in 'quality and relevance'. That evidence is weak for the later claims about accurate diagnostic personalization, but weakness of evidence is not circularity: the LLM outputs are not used as the ground truth that defines the claimed capability, and no parameter is fitted and then renamed a prediction. The paper itself frames the work as 'preliminary explorations' and acknowledges deployment hurdles, including a footnote that 'there is still a significant hurdle in achieving those requirements' and a Section 4 list of 'limitations and challenges'. The only potentially self-referential element is the citation of prior work by the authors (e.g., [63]) when defining pedagogical agents, but that background claim is corroborated by independent meta-analyses ([12], [54], [60]) and is not load-bearing for the neurosymbolic-AI-specific thesis. No self-definitional equivalence, imported uniqueness theorem, ansatz-smuggling-via-citation, or renaming of a known result is present. Therefore the paper exhibits no significant circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

The central claims rest on unvalidated assumptions about LLM reasoning accuracy, knowledge-graph completeness, and the transfer of pedagogical agent benefits to a neurosymbolic architecture. The paper provides no controlled experiments, no quantitative measures, and no independent benchmark, so the ledger is dominated by domain assumptions and ad-hoc-to-paper premises.

assumptions (3)
  • ad hoc to paper LLMs augmented with knowledge graphs can reason accurately enough to diagnose learner knowledge states
    Assumed in Sections 3.1 and 3.2, and implicitly in the exploratory studies of Section 4; no evidence demonstrates that retrieval-augmented LLMs can reliably detect knowledge gaps.
  • domain assumption Pedagogical agents improve learning and motivation as claimed by prior literature
    The paper relies on prior meta-analyses (Castro-Alonso et al. 2021; Schroeder et al. 2013) to justify the PA component; this is a reasonable domain assumption from cited work, but the transfer to an NAI-based agent is untested.
  • ad hoc to paper The proposed system can generate multimodal, multilingual content at scale without degrading quality
    Stated as a capacity in Sections 3.2 and 4, but no evaluation of audio, video, translation, or accessibility outputs is provided.
invented entities (1)
  • NaPA (Neuro-symbolic AI-augmented Pedagogical Agent)
    purpose: An embodied pedagogical agent that combines neural LLMs, symbolic knowledge graphs, and evidence-based pedagogy to deliver adaptive, multimodal, personalized instruction.
    The paper proposes the NaPA as a conceptual architecture but provides no implementation or falsifiable prediction that could validate its existence or effectiveness outside the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Education in the Era of Neurosymbolic AI." pith.science (2026). https://pith.science/paper/G6I4EEGY

@misc{pith2026241112763,
  author       = {Pith},
  title        = {Pith review of: Education in the Era of Neurosymbolic AI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/G6I4EEGY}},
  note         = {Machine review of arXiv:2411.12763}
}
read the original abstract

Education is poised for a transformative shift with the advent of neurosymbolic artificial intelligence (NAI), which will redefine how we support deeply adaptive and personalized learning experiences. NAI-powered education systems will be capable of interpreting complex human concepts and contexts while employing advanced problem-solving strategies, all grounded in established pedagogical frameworks. This will enable a level of personalization in learning systems that to date has been largely unattainable at scale, providing finely tailored curricula that adapt to an individual's learning pace and accessibility needs, including the diagnosis of student understanding of subjects at a fine-grained level, identifying gaps in foundational knowledge, and adjusting instruction accordingly. In this paper, we propose a system that leverages the unique affordances of pedagogical agents -- embodied characters designed to enhance learning -- as critical components of a hybrid NAI architecture. To do so, these agents can thus simulate nuanced discussions, debates, and problem-solving exercises that push learners beyond rote memorization toward deep comprehension. We discuss the rationale for our system design and the preliminary findings of our work. We conclude that education in the era of NAI will make learning more accessible, equitable, and aligned with real-world skills. This is an era that will explore a new depth of understanding in educational tools.

Figures

Figures reproduced from arXiv: 2411.12763 by the authors.

Figure 1
Figure 1. An overview of the iterative learning-intervention cycle enabled by the integration of Neurosymbolic AI in education. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 57 canonical work pages

  1. [1]

    O. O. Adesope, D. A. Trevisan, and N. Sundararajan. Re- thinking the use of tests: A meta-analysis of practice test- ing. Review of educational research, 87(3):659–701, 2017

  2. [2]

    P. K. Agarwal, L. D. Nunes, and J. R. Blunt. Retrieval practice consistently benefits student learning: A system- atic review of applied research in schools and classrooms. Educational Psychology Review, 33(4):1409–1453, 2021

  3. [3]

    Q. U. Ain, M. A. Chatti, K. G. C. Bakar, S. Joarder, and R. Alatrash. Automatic construction of educational knowl- edge graphs: a word embedding-based approach. Informa- tion, 14(10):526, 2023

  4. [4]

    Q. U. Ain, M. A. Chatti, P. A. Meteng Kamdem, R. Ala- trash, S. Joarder, and C. Siepmann. Learner modeling and recommendation of learning resources using personal knowledge graphs. In Proceedings of the 14th Learn- ing Analytics and Knowledge Conference, pages 273–283, 2024

  5. [5]

    R. K. Atkinson. Optimizing learning from examples us- ing animated pedagogical agents. Journal of Educational Psychology, 94(2):416, 2002

  6. [6]

    Balog and T

    K. Balog and T. Kenter. Personal knowledge graphs: A research agenda. In Proceedings of the 2019 ACM SI- GIR International Conference on Theory of Information Retrieval, pages 217–220, 2019

  7. [7]

    Barua, G

    A. Barua, G. Brase, K. Dong, P. Hitzler, and E. Y . Vasser- man. On the psychology of GPT-4: moderately anx- ious, slightly masculine, honest, and humble. CoRR, abs/2402.01777, 2024

  8. [8]

    T. B. Brown. Language models are few-shot learners. arXiv preprint arXiv:2005.14165, 2020

Show all 64 references
  1. [9]

    Brunmair and T

    M. Brunmair and T. Richter. Similarity matters: A meta- analysis of interleaved learning and its moderators. Psy- chological bulletin, 145(11):1029, 2019

  2. [10]

    Brusilovsky

    P. Brusilovsky. Student model centered architecture for in- telligent learning environments. In Proceedings of Fourth international conference on User Modeling, pages 15–19, 1994

  3. [11]

    S. K. Carpenter, S. C. Pan, and A. C. Butler. The science of effective learning with spacing and retrieval practice. Nature Reviews Psychology, 1(9):496–511, 2022

  4. [12]

    J. C. Castro-Alonso, R. M. Wong, O. O. Adesope, and F. Paas. Effectiveness of multimedia pedagogical agents predicted by diverse theories: A meta-analysis. Educa- tional Psychology Review, 33:989–1015, 2021

  5. [13]

    B. Chen, Z. Zhang, N. Langrené, and S. Zhu. Un- leashing the potential of prompt engineering in large lan- guage models: a comprehensive review. arXiv preprint arXiv:2310.14735, 2023

  6. [14]

    P. Chen, Y . Lu, V . W. Zheng, X. Chen, and B. Yang. Knowedu: A system to construct knowledge graph for ed- ucation. Ieee Access, 6:31553–31563, 2018

  7. [15]

    Clarebout, J

    G. Clarebout, J. Elen, W. L. Johnson, and E. Shaw. Animated pedagogical agents: An opportunity to be grasped? Journal of Educational multimedia and hyper- media, 11(3):267–286, 2002

  8. [16]

    Firth, I

    J. Firth, I. Rivers, and J. Boyle. A systematic review of interleaving as a concept learning strategy. Review of Ed- ucation, 9(2):642–684, 2021

  9. [17]

    Y . Gao, Y . Xiong, X. Gao, K. Jia, J. Pan, Y . Bi, Y . Dai, J. Sun, M. Wang, and H. Wang. Retrieval-augmented generation for large language models: A survey. arXiv preprint arXiv:2312.10997, 2023

  10. [18]

    A. d. Garcez and L. C. Lamb. Neurosymbolic ai: The 3 rd wave. Artificial Intelligence Review, 56(11):12387–12406, 2023

  11. [19]

    Gligorea, M

    I. Gligorea, M. Cioca, R. Oancea, A.-T. Gorski, H. Gorski, and P. Tudorache. Adaptive learning using artificial intel- ligence in e-learning: a literature review. Education Sci- ences, 13(12):1216, 2023

  12. [20]

    https://openai.com/index/hello-gpt-4o/

    GPT-4o. https://openai.com/index/hello-gpt-4o/

  13. [21]

    J. M. Harley, F. Bouchet, M. S. Hussain, R. Azevedo, and R. Calvo. A multi-componential analysis of emotions dur- ing complex learning with an intelligent multi-agent sys- 8 tem. Computers in Human Behavior, 48:615–625, 2015

  14. [22]

    Heidig and G

    S. Heidig and G. Clarebout. Do pedagogical agents make a difference to student motivation and learning? Educa- tional Research Review, 6(1):27–54, 2011

  15. [23]

    P. Hitzler. A review of the semantic web field. Commun. ACM, 64(2):76–83, 2021

  16. [24]

    Hitzler and M

    P. Hitzler and M. K. Sarker. Neuro-symbolic artificial in- telligence: The state of the art. IOS press, 2022

  17. [25]

    Hogan, E

    A. Hogan, E. Blomqvist, M. Cochez, C. d’Amato, G. D. Melo, C. Gutierrez, S. Kirrane, J. E. L. Gayo, R. Navigli, S. Neumaier, et al. Knowledge graphs. ACM Computing Surveys (Csur), 54(4):1–37, 2021

  18. [26]

    B. Hu, L. Zheng, J. Zhu, L. Ding, Y . Wang, and X. Gu. Teaching plan generation and evaluation with gpt-4: Un- leashing the potential of llm in instructional design. IEEE Transactions on Learning Technologies, 2024

  19. [27]

    Huang, J

    C.-Y . Huang, J. Wei, and T.-H. K. Huang. Generating edu- cational materials with different levels of readability using llms. In Proceedings of the Third Workshop on Intelligent and Interactive Writing Assistants, pages 16–22, 2024

  20. [28]

    Huynh, C

    J. Huynh, C. Jiao, P. Gupta, S. Mehri, P. Bajaj, V . Chaud- hary, and M. Eskenazi. Understanding the effectiveness of very large language models on dialog evaluation. arXiv preprint arXiv:2301.12004, 2023

  21. [29]

    E. Ilkou. Personal knowledge graphs: Use cases in e- learning platforms. In Companion Proceedings of the Web Conference 2022, pages 344–348, 2022

  22. [30]

    Ilkou, T

    E. Ilkou, T. Tolmachova, M. Fisichella, and D. Taibi. Col- labgraph: A graph-based collaborative search summary vi- sualization. IEEE Trans. Learn. Technol., 16(3):382–398, 2023

  23. [31]

    C. Jeong. Fine-tuning and utilization methods of domain- specific llms. arXiv preprint arXiv:2401.02981, 2024

  24. [32]

    Jiang, M

    B. Jiang, M. Gu, and Y . Du. Recent advances in intelli- gent textbooks for better learning. Learning: Designing the Future, page 247, 2023

  25. [33]

    S. H. Kang. Spaced repetition promotes efficient and ef- fective learning: Policy implications for instruction.Policy Insights from the Behavioral and Brain Sciences, 3(1):12– 19, 2016

  26. [34]

    J. D. Karpicke and J. R. Blunt. Retrieval practice produces more learning than elaborative studying with concept map- ping. Science, 331(6018):772–775, 2011

  27. [35]

    Kojima, S

    T. Kojima, S. S. Gu, M. Reid, Y . Matsuo, and Y . Iwa- sawa. Large language models are zero-shot reason- ers. Advances in neural information processing systems , 35:22199–22213, 2022

  28. [36]

    Latimier, H

    A. Latimier, H. Peyre, and F. Ramus. A meta-analytic review of the benefit of spacing out retrieval practice episodes on retention. Educational Psychology Review , 33:959–987, 2021

  29. [37]

    Lewis, E

    P. Lewis, E. Perez, A. Piktus, F. Petroni, V . Karpukhin, N. Goyal, H. Küttler, M. Lewis, W.-t. Yih, T. Rocktäschel, et al. Retrieval-augmented generation for knowledge- intensive nlp tasks. Advances in Neural Information Pro- cessing Systems, 33:9459–9474, 2020

  30. [38]

    T. Li, G. Zhang, Q. D. Do, X. Yue, and W. Chen. Long- context llms struggle with long in-context learning. arXiv preprint arXiv:2404.02060, 2024

  31. [39]

    L. Liu, H. Du, X. Zhang, M. Guo, H. Wang, and M. Wang. A question-answering assistant over personal knowledge graph. In Proceedings of the 47th International ACM SI- GIR Conference on Research and Development in Infor- mation Retrieval, pages 2708–2712, 2024

  32. [40]

    P. Liu, W. Yuan, J. Fu, Z. Jiang, H. Hayashi, and G. Neu- big. Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing. ACM Computing Surveys, 55(9):1–35, 2023

  33. [41]

    https://ai.meta.com/blog/llama-3-2-connect- 2024-vision-edge-mobile-devices/

    Llama 3.2. https://ai.meta.com/blog/llama-3-2-connect- 2024-vision-edge-mobile-devices/

  34. [42]

    R. E. Mayer, K. Sobko, and P. D. Mautone. Social cues in multimedia learning: Role of speaker’s voice. Journal of educational Psychology, 95(2):419, 2003

  35. [43]

    Murray, M

    R. Murray, M. Shea, B. Shea, and R. Harlin. Issues in edu- cation: Avoiding the one-size-fits-all curriculum: Textsets, inquiry, and differentiating instruction. Childhood educa- tion, 81(1):33–35, 2004

  36. [44]

    https://github.com/kastle-lab/EduNAILearning-Research

    Nai in education exploratory research. https://github.com/kastle-lab/EduNAILearning-Research. Accessed: 2024-11-16

  37. [45]

    J. C. Nesbit, O. O. Adesope, Q. Liu, and W. Ma. How ef- fective are intelligent tutoring systems in computer science education? In 2014 IEEE 14th international conference on advanced learning technologies , pages 99–103. IEEE, 2014

  38. [46]

    N. F. Noy, Y . Gao, A. Jain, A. Narayanan, A. Patterson, and J. Taylor. Industry-scale knowledge graphs: lessons and challenges. Commun. ACM, 62(8):36–43, 2019

  39. [47]

    https://laptop.org/

    One Laptop Per Child. https://laptop.org/

  40. [48]

    Prather, P

    J. Prather, P. Denny, J. Leinonen, B. A. Becker, I. Albluwi, M. Craig, H. Keuning, N. Kiesler, T. Kohn, A. Luxton- Reilly, et al. The robots are here: Navigating the generative ai revolution in computing education. In Proceedings of the 2023 Working Group Reports on Innovation...

  41. [49]

    https://projectconnect.unicef.org/map

    Project connect unicef. https://projectconnect.unicef.org/map

  42. [50]

    K. Qu, K. C. Li, B. T. Wong, M. M. Wu, and M. Liu. A survey of knowledge graph approaches and applications in education. Electronics, 13(13):2537, 2024

  43. [51]

    Reeves and C

    B. Reeves and C. Nass. The media equation: How people treat computers, television, and new media like real people. Cambridge, UK, 10(10):19–36, 1996

  44. [52]

    H. L. Roediger and A. C. Butler. The critical role of re- trieval practice in long-term retention. Trends in cognitive sciences, 15(1):20–27, 2011

  45. [53]

    Schneider, M

    S. Schneider, M. Beege, S. Nebel, L. Schnaubert, and G. D. Rey. The cognitive-affective-social theory of learning in digital environments (castle). Educational Psychology Re- view, 34(1):1–38, 2022

  46. [54]

    N. L. Schroeder, O. O. Adesope, and R. B. Gilbert. How effective are pedagogical agents for learning? a meta- analytic review. Journal of Educational Computing Re- search, 49(1):1–39, 2013

  47. [55]

    N. L. Schroeder and C. M. Gotch. Persisting issues in ped- agogical agent research. Journal of Educational Comput- ing Research, 53(2):183–204, 2015

  48. [56]

    R. F. Siegle, N. L. Schroeder, H. C. Lane, and S. D. Craig. Twenty-five years of learning with pedagogical agents: History, barriers, and opportunities. TechTrends, 67(5):851–864, 2023

  49. [57]

    M. G. Skjæveland, K. Balog, N. Bernard, W. Łajewska, and T. Linjordet. An ecosystem for personal knowledge graphs: A survey and research roadmap. AI Open, 5:55– 69, 2024

  50. [58]

    https://sdgs

    Sustainability Development Goals. https://sdgs. un.org/goals/goal4. Accessed: 2024-11-16

  51. [59]

    K. VanLehn. The relative effectiveness of human tutor- ing, intelligent tutoring systems, and other tutoring sys- tems. Educational psychologist, 46(4):197–221, 2011

  52. [60]

    Y . Wang, S. Gong, Y . Cao, Y . Lang, and X. Xu. The effects of affective pedagogical agent in multimedia learning envi- ronments: A meta-analysis. Educational Research Review, 38:100506, 2023

  53. [61]

    White, Q

    J. White, Q. Fu, S. Hays, M. Sandborn, C. Olea, H. Gilbert, A. Elnashar, J. Spencer-Smith, and D. C. Schmidt. A prompt pattern catalog to enhance prompt engineering with chatgpt. arXiv preprint arXiv:2302.11382, 2023. 9

  54. [62]

    C. Wu, B. Jing, X. Gong, and X. Ma. The zoomorphic effect: A contribution to the study of images of pedagog- ical agents for children’s learning in instructional videos. Journal of Computer Assisted Learning, 39(5):1620–1635, 2023

  55. [63]

    Zhang, C

    S. Zhang, C. D. Jaldi, N. L. Schroeder, A. A. López, J. R. Gladstone, and S. Heidig. Pedagogical agent design for k-12 education: A systematic review. Computers & Edu- cation, page 105165, 2024

  56. [64]

    H. Zhao, H. Chen, F. Yang, N. Liu, H. Deng, H. Cai, S. Wang, D. Yin, and M. Du. Explainability for large lan- guage models: A survey. ACM Transactions on Intelligent Systems and Technology, 15(2):1–38, 2024. 10

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.