An Experience Report on a Pedagogically Controlled, Curriculum-Constrained AI Tutor for SE Education

Anne Koziolek; Dominik Fuch{\ss}; Kai Marquardt; Luca H\"uttner; Lucia Happe

REVIEW 3 major objections 6 minor 30 references

An Experience Report on a Pedagogically Controlled, Curriculum-Constrained AI Tutor for SE Education

T0 review · 3 major / 6 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that a general-purpose language model, constrained by a tagged knowledge base and prompt rules, can tutor secondary computing students safely and acceptably without any model training.

desk verdict Honest, modest pilot of a no-fine-tuning GPT-4 tutor with a tagged knowledge base; the safety claims outrun the evidence, but the design blueprint and candid reporting make it worth a serious look. read the letter →

arxiv 2512.11882 v1 pith:SNBPQGTC submitted 2025-12-08 cs.CY cs.AIcs.HCcs.SE

Lucia Happe , Dominik Fuch{ss} , Luca Hüttner , Kai Marquardt , Anne Koziolek This is my paper · ORCID

classification cs.CYcs.AIcs.HCcs.SE

keywords AItutorslargelanguagemodelschatbotspromptengineeringcurriculumalignmentknowledgebasetaggingscaffoldingsoftwareeducation

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

This paper tries to establish that a general-purpose large language model can be turned into a safe, curriculum-aligned tutor for secondary computing students using only structure and prompting—no custom training. The method combines a modular knowledge base in which each course fragment is tagged by pedagogical role (task, hint, explanation, do-not-show solution, motivation, misconception) with an instruction set that restricts the model to that content and forbids direct solutions. The authors report a pilot study with 13 students and teachers in which perceived usefulness, ease of use, and intention to use were rated positively, with ease-of-use items scoring highest. If the central claim holds, schools could deploy personalized, scaffolded tutoring on top of existing language-model APIs without expensive fine-tuning, and teachers could inspect and control exactly what the tutor is allowed to say.

What carries the argument

The load-bearing mechanism is the tagged Markdown knowledge base plus the instruction set. Each content unit carries a semantic tag that tells the tutor how to use it: prompts for tasks trigger hint-first scaffolding; explanations support answer-then-analogy patterns; solution sections are explicitly marked 'do not show' and the prompt forbids their release. The tutor's retrieval is restricted to these files via a file-grounding search tool, and a layered traversal strategy escalates from hints to explanations only as needed. Thread memory lets it refer to earlier turns. This combination—rather than any model training—is what the authors say constrains behavior.

What would settle it

Feed the tutor a set of off-curriculum questions (e.g., 'What is the capital of France?') and direct requests for solution sections; if it answers the off-topic questions or leaks a tagged solution in a nontrivial share of trials, the claim that the architecture eliminates hallucinations and enforces curriculum constraints is falsified.

Watch

Extended reading notes

Core claim

The paper claims that a general-purpose LLM can be made into a safe, curriculum-aligned tutor for secondary computing students by giving it two things: a modular knowledge base in which every piece of course content is tagged by pedagogical role (task, hint, explanation, solution marked do-not-show, motivation, misconception), and a system prompt that instructs the model to use only that knowledge base, to answer only course questions, to speak in age-appropriate language, and never to reveal tagged solutions. The authors argue that this combination, using the model's file-grounded retrieval and thread memory, yields 'pedagogical safety' and eliminates hallucinations without any custom train

Load-bearing premise

The load-bearing premise is that a general-purpose LLM, when pointed at a tagged knowledge base and instructed to stay inside it, will actually do so—staying on-curriculum, refusing to solve 'do not show' tasks, and never hallucinating—under real student questioning.

Editorial extensions

If this is right

Teachers and course authors can extend the tutor to new topics simply by adding new tagged files; no retraining or model access changes are needed.
Students get low-stakes, always-available, scaffolded guidance that withholds final answers, encouraging them to work through problems themselves.
The same architecture can be replicated for other courses and age groups by re-tagging content and adjusting the prompt's language level.
Because the knowledge base is inspectable, educators can verify exactly what content the tutor is allowed to use, supporting trust and accountability.
Positive acceptance results, if they hold with the intended younger cohort, support a complementary model where AI tutors handle routine individual questions and free teachers for higher-value interaction.

Reading between the lines

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

The paper's strongest safety claim—that hallucinations are eliminated—is architecture-level, but the reported evaluation only measures self-reported acceptance; a direct adversarial test with off-topic questions and solution-extraction attempts is the natural next step.
The acceptance data come from university-age learners and pre-service teachers, so transferring the numbers to 14- to 16-year-olds is an assumption the authors flag; a replication with the target age group would settle it.
The tagged-content pattern is not specific to computing: any school subject whose material can be chunked into tasks, hints, explanations, and misconceptions could benefit from the same constraint-by-tagging design.
The fixed 'never reveal solutions' policy may need to become adaptive, because in some contexts (e.g., exam review or advanced students) showing a worked example is pedagogically appropriate; the paper doesn't explore this flexibility.

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.

Desk Editor's Note

read the letter

You should know two things. First, this is a genuinely useful design note for building curriculum-tethered GPT tutors without fine-tuning. Second, the paper repeatedly claims the architecture “eliminates hallucinations” and guarantees curriculum fidelity, but nothing in the evaluation actually measures that behavior.

What is new: the specific combination — GPT-4o via the Assistant API, file_search over a semantically tagged Markdown knowledge base, a system prompt that forbids direct solutions, and a hint-first traversal strategy — is a real but modest design contribution. It is cheap, transparent, and does not require custom training. The system description and prompt are clear, and the authors include enough detail to reproduce the approach. That is more than many AI-ed papers offer.

The empirical core is honest but thin. Thirteen adults (six pre-service teachers, seven university students) filled out a TAM questionnaire after 30–45 minutes. Item means, standard deviations, per-item tables, and representative open-ended quotes are all reported. The Limitations section is unusually candid: they concede the sample size, the age mismatch, the prototype status, the absence of interaction logs, and the lack of longitudinal data. That counts for something.

The load-bearing problem is that §§3.3 and 3.7 assert the system “eliminates hallucinations” and ensures “pedagogical safety” as established fact, while the evaluation measures only self-reported acceptance. There is no adherence audit, no off-topic probe, no solution-leakage test, and no baseline or control condition. The qualitative feedback already hints at constraint failures — “It is sometimes unclear what the tutor’s level of knowledge is,” and responses were “too generic or overly cautious.” The stress-test note is right about this: the safety guarantee is unsupported. A second gap is the transfer from 23.7-year-old learners to the intended 14–16-year-olds; the authors flag it, but it remains unresolved.

These are evidentiary gaps, not integrity problems. The central claim as framed in the title and abstract — that a no-fine-tuning GPT-4 tutor grounded in a tagged knowledge base reliably stays inside curriculum boundaries — is plausible but not demonstrated. The paper would be stronger if it either softened the hallucination language or added a behavioral constraint measure.

Who is this for? People building AI tutors for K-12 or low-resource classrooms will get practical value from the architecture walkthrough and the candid discussion of pitfalls. I would send it to peer review: it is a legitimate experience report with a reusable design, and referees can push for the missing behavioral evidence. The revision bar is clear — add a constraint test or dial back the claims.

Referee Report

3 major / 6 minor

Summary. The paper reports on the design and pilot evaluation of RockStartIT Tutor, a GPT-4o-based assistant built on the OpenAI Assistant API with a modular, semantically tagged Markdown knowledge base and a hand-crafted system prompt. The authors claim that this architecture, without fine-tuning, provides curriculum-constrained, pedagogically safe tutoring that 'eliminates hallucinations' (§3.3, §3.7). They evaluate the system with a TAM-based questionnaire and open-ended questions from 13 participants (7 learners with mean age 23.7 and 6 teachers with mean age 32.5). The reported quantitative results are generally positive: PU/PEOU/IU item means range from about 3.4 to 4.6 on a 5-point scale, and qualitative feedback indicates appreciation for low-stakes, scaffolded support. The paper positions the work as an experience report and acknowledges several limitations, including small sample size, target-age mismatch, missing interaction logs, and lack of longitudinal data.

Significance. If the central architectural claim were substantiated, the contribution would be valuable: a no-fine-tuning, prompt-and-knowledge-base approach to curriculum-constrained tutoring is lightweight, replicable, and directly relevant to a gap in the literature on AI tutors for secondary computing education. The paper also has strengths: the evaluation data are reported item-by-item with means, standard deviations, and participant counts; the limitations section is unusually candid; and the qualitative quotes give a concrete sense of user perceptions. However, the strongest advertised result—that the system 'eliminates hallucinations and ensures pedagogical safety'—is not measured by the evaluation, and the acceptance data come from adults rather than the intended 14–16-year-old audience. As a consequence, the paper currently supports a modest claim about perceived usability and usefulness in a small adult pilot, not the safety or effectiveness claims in the abstract and Section 3.

major comments (3)

[§3.3, §3.7, §6; Evaluation §4] The paper asserts that the architecture 'eliminates hallucinations and ensures pedagogical safety' (§3.3) and that responses are 'grounded in a predefined, structured knowledge base, eliminating hallucinations' (§3.7). No behavioral evidence is reported to support this. The evaluation measures self-reported TAM perceptions only; there is no adherence test, no hallucination audit, no probe of off-topic answers, and no test of whether 'do not show' solutions leak. The Limitations section (§6) even states that 'no behavioral usage data were collected' and that trust calibration was not assessed. The authors' own qualitative data hint at uncontrolled behavior: 'It is sometimes unclear what the tutor’s level of knowledge is' and responses 'too generic or overly cautious' (§4.3.4). This is a load-bearing gap: a system can be perceived as useful and easy to use while still hallucinating or leak
[§4.2, Table 1; §7] The intended target group is secondary school students aged 14–16 (§3.1), but the evaluation used 7 learners with mean age 23.7 and 6 teachers with mean age 32.5. The paper acknowledges this mismatch in §6, yet the conclusion states that 'we observed high levels of perceived usefulness, usability, and intention to use' and that 'the positive reception from both students and educators suggests a viable role' for future SE education. Given that the authors themselves attribute some learner ratings (PU4, PU5, M = 3.43) to the difference between the university-student sample and the intended audience, the central acceptance claim cannot be generalized to the target population. I recommend reframing the conclusions to explicitly restrict them to the adult pilot participants, or providing additional data from the intended age group. This is a required change because the title and abstract pres
[§4.3, Table 3] Several correlational claims are made from a very small sample: PU1–PU4 r=0.77, PU2–PU4 r=0.58, PEOU4–IU2 r=0.79, and age–IU3 r=0.48 among teachers only. With n=13 overall and n=6 for teacher-only correlations, no significance tests or confidence intervals are reported, and Spearman versus Pearson is not specified. Statements such as 'a strong positive correlation' and 'a moderate positive correlation' give these exploratory associations more inferential weight than they can bear. This is secondary to the main claims but contributes to the impression of stronger evidence than the data provide. I suggest either removing the correlational analysis or explicitly labelling it as descriptive, with a caution about the small sample and multiple comparisons.

minor comments (6)

[Throughout] Terminology is inconsistent: the system is called 'AI Tutor', 'KI Tutor', and 'RockStartIT Tutor'; 'RockStart IT' and 'RockStartIT' are both used. Please standardize.
[Table 3] PEOU5 has N=12 while all other items have N=13; the missing response is not explained. This should be footnoted or clarified.
[Figures 3–5] The stacked bar charts lack a clear legend or explicit mapping from colors to the Likert labels. It is also unclear how counts align with the means in Table 3; please make the figures self-contained.
[§4.3.1] The text says 'PU3 received the highest level of agreement', but Table 3 shows PU3 mean 3.69, equal to PU1 and PU5; the claim may refer to a specific response distribution rather than the mean, but as written it is confusing.
[§4.3.4] The open-ended data are presented as thematic summaries, but the analysis procedure (coding, inter-rater reliability, or even a single-coder thematic analysis) is not described. Since the qualitative results are part of the mixed-methods evidence, a sentence on how themes were derived would strengthen the report.
[References] Several references are incomplete or inconsistent, e.g., [12] includes 'Submitted to EAIT (2023)' as if it were a publication venue, [29] gives only 'In ACM SIGCSE', and 'Holstein and et al.' appears in the text. Please normalize.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is an empirical experience report; the architecture-to-safety claims are unsupported rather than definitionally forced.

full rationale

The paper does not derive any quantitative result from a fitted parameter or equation. Its central claims are architectural (grounding and prompting constrain GPT-4) and perceptual (TAM acceptance). The §3.3/§3.7 assertion that the Assistant API file_search plus prompt 'eliminates hallucinations' is an unsupported empirical claim, not a circular one: nothing in the paper defines the system's safety in terms of the measured TAM outcomes, and the TAM items do not test adherence. The only self-citations ([9][10][11][12]) motivate the RockStartIT context and prior course design; they are not used to justify the tutor acceptance results or the hallucination claim. The Limitations section explicitly acknowledges the threats to the acceptance claim (n=13, wrong age group, no behavioral logs, no trust calibration), which is evidence of non-circular honest reporting. No fitted input is relabeled as a prediction, no uniqueness theorem is imported, and no known result is renamed. Therefore no specific reduction to inputs can be exhibited; score 0.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

This is an architectural/empirical paper, not a derivational one, so the ledger records hand-tuned design choices and background assumptions rather than fitted constants. The two hand-tuned artifacts (system prompt; tag set plus traversal rules) carry the behavior-control claim. The most consequential item is the unmeasured grounding assumption — that file_search plus a prompt "eliminates hallucinations" (§3.3, §3.7) — which is asserted and never evaluated. No invented entities: the AI Tutor/KI Tutor is the software artifact under test, though its status shifts between "fully functional" (§2.6) and "early-stage prototype" (§6). No free numerical parameters.

free parameters (2)

Final system prompt and instruction set (Prompt 1) = Prompt 1 text in §3.4; described as the product of "iterative prompt tuning" (§3.4)
The behavior-defining prompt was hand-tuned by the authors to make the model scaffold, refuse solutions, and stay on-curriculum; no ablation or robustness analysis shows how sensitive the claimed behavior is to this exact wording.
Knowledge-base tagging scheme and traversal rules = Tag set {tasks, hints, explanations, solutions, motivation, misconceptions} and hint-first escalation order, §3.3
The semantic tags and the layered traversal are design choices adjusted to the course and the authors' pedagogical judgment; they are not derived from any theory or benchmark, and their contribution to the outcomes is untested.

assumptions (4)

domain assumption File_search grounding plus the system prompt reliably prevents GPT-4o from answering outside the uploaded curriculum content and from revealing "do not show" solutions.
Invoked in §3.3 ("This eliminates hallucinations and ensures pedagogical safety") and §3.4; asserted as fact but never measured in §4 — no adherence, hallucination, or solution-leakage test is reported.
domain assumption Self-reported TAM acceptance by 13 adults is a valid proxy for acceptance and educational value for the intended 14–16-year-old secondary audience.
Implicit throughout §§4–5; the age mismatch is disclosed in §6, but the generalization is load-bearing for the abstract's and conclusion's framing.
domain assumption The tagged Markdown knowledge base is complete and pedagogically correct for the RockStartIT module, so that confining the model to it yields good tutoring.
Stated in the §3.1–3.3 design requirements; no content validation, coverage check, or expert-content audit is reported.
domain assumption Thread-based memory via the Assistant API works as described across multi-turn interactions and shapes scaffolding as intended.
Described in §3.5 (Figure 2); no evaluation data on thread behavior or context handling is provided.

how reviews work

0 comments

Cite this review

Pith. "Pith review of An Experience Report on a Pedagogically Controlled, Curriculum-Constrained AI Tutor for SE Education." pith.science (2026). https://pith.science/paper/SNBPQGTC

@misc{pith2026251211882,
  author       = {Pith},
  title        = {Pith review of: An Experience Report on a Pedagogically Controlled, Curriculum-Constrained AI Tutor for SE Education},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SNBPQGTC}},
  note         = {Machine review of arXiv:2512.11882}
}

read the original abstract

The integration of artificial intelligence (AI) into education continues to evoke both promise and skepticism. While past waves of technological optimism often fell short, recent advances in large language models (LLMs) have revived the vision of scalable, individualized tutoring. This paper presents the design and pilot evaluation of RockStartIT Tutor, an AI-powered assistant developed for a digital programming and computational thinking course within the RockStartIT initiative. Powered by GPT-4 via OpenAI's Assistant API, the tutor employs a novel prompting strategy and a modular, semantically tagged knowledge base to deliver context-aware, personalized, and curriculum-constrained support for secondary school students. We evaluated the system using the Technology Acceptance Model (TAM) with 13 students and teachers. Learners appreciated the low-stakes environment for asking questions and receiving scaffolded guidance. Educators emphasized the system's potential to reduce cognitive load during independent tasks and complement classroom teaching. Key challenges include prototype limitations, a small sample size, and the need for long-term studies with the target age group. Our findings highlight a pragmatic approach to AI integration that requires no model training, using structure and prompts to shape behavior. We position AI tutors not as teacher replacements but as enabling tools that extend feedback access, foster inquiry, and support what schools do best: help students learn.

Figures

Figures reproduced from arXiv: 2512.11882 by the authors.

**Figure 2.** Threaded tutor interaction from RockStartIT’s bee [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Perceived Usefulness (PU) ratings from teachers / [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Perceived Ease of Use (PEOU) ratings [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Intention to Use (IU) ratings from teachers / learners [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 1 canonical work pages

[1]

2013.International handbook of metacognition and learning technologies

Roger Azevedo and Vincent Aleven. 2013.International handbook of metacognition and learning technologies. Springer. 1000 pages. doi:10.1007/978-1-4419-5546-3

work page doi:10.1007/978-1-4419-5546-3 2013
[2]

Narges Bakhtiar and Sidney K. D’Mello. 2022. A Framework for Trust in Educa- tional AI.International Journal of Artificial Intelligence in Education32 (2022), 1–23

2022
[3]

Patrick Bassner, Eduard Frankford, and Stephan Krusche. 2024. Iris: An AI- Driven Virtual Tutor for Computer Science Education. InProceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1(Milan, Italy)(ITiCSE 2024). Association for Computing Machinery, New York, NY, USA, 394–400. doi:10.1145/3649217.3653543

arXiv 2024
[4]

Michelene TH Chi, Marguerite Roy, and Robert G Hausmann. 2008. Observing tutorial dialogues collaboratively: Insights about human tutoring effectiveness from vicarious learning.Cognitive science32, 2 (3 2008), 301–341

2008
[5]

2001.Oversold and Underused: Computers in the Classroom

Larry Cuban. 2001.Oversold and Underused: Computers in the Classroom. Harvard University Press

2001
[6]

Wei Dai, Jionghao Lin, Hua Jin, Tongguang Li, Yi-Shan Tsai, Dragan Gašević, and Guanliang Chen. 2023. Can Large Language Models Provide Feedback to Students? A Case Study on ChatGPT. In2023 IEEE International Conference on Advanced Learning Technologies (ICALT). 323–325. doi:10.1109/ICALT58122.2023. 00100

arXiv 2023
[7]

Fred D. Davis. 1989. Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology.MIS Quarterly13, 3 (9 1989), 319–340. doi:10.2307/249008 Publisher: Management Information Systems Research Center, University of Minnesota

doi:10.2307/249008 1989
[8]

Eduard Frankford, Clemens Sauerwein, Patrick Bassner, Stephan Krusche, and Ruth Breu. 2024. AI-Tutoring in Software Engineering Education, In Proceedings of the 46th International Conference on Software Engineering: Software Engineer- ing Education and Training (Lisbon Portugal, 2024-04-14).2024 IEEE/ACM 46th International Conference on Software Engineeri...

arXiv 2024

Show all 30 references

[9]

Philip Gross and Lucia Happe. 2022. Challenges and Opportunities of AI Tutors in Software Engineering Education.Informatics in Education(2022)

2022
[10]

Lucia Happe and Barbora Buhnova. 2022. Frustrations Steering Women Away From Software Engineering.IEEE Software39, 4 (7 2022), 63–69. doi:10.1109/MS. 2021.3099077

2022
[11]

Lucia Happe, Barbora Buhnova, Anne Koziolek, and Ingo Wagner. 2021. Effective measures to foster girls’ interest in secondary computer science education.Edu- cation and Information Technologies26, 3 (2021), 2811–2829. doi:10.1007/s10639- 020-10379-x

2021 doi
[12]

Lucia Happe and et al. 2025. Authentic Interdisciplinary Online Courses for Alter- native Pathways into Computer Science. doi:10.1016/j.jss.2024.112240 Submitted to EAIT (2023)

2025
[13]

Arto Hellas, Juho Leinonen, Sami Sarsa, Charles Koutcheme, Lilja Kujanpää, and Juha Sorva. 2023. Exploring the Responses of Large Language Models to Beginner Programmers’ Help Requests. InProceedings of the 2023 ACM Conference on International Computing Education Research - Vo...

2023
[14]

Wayne Holmes et al. 2019. Artificial intelligence in education: Promises and implications for teaching and learning.OECD Publishing(2 2019). https://oro. open.ac.uk/60255/

2019
[15]

Kenneth Holstein and et al. 2019. Student perceptions of AI tutors in the classroom. InCHI ’19

2019
[16]

Greg Kestin, Kelly Miller, Anna Klales, Timothy Milbourne, and Gregorio Ponti
[17]

Daniel Lee and Edward Palmer. 2025. Prompt engineering in higher education: a systematic review to help inform curricula.International Journal of Educational Technology in Higher Education22, 1 (2025), 7

2025
[18]

John D Lee and Katrina A See. 2004. Trust in automation: Designing for appro- priate reliance.Human Factors46, 1 (3 2004), 50–80. https://www.engineering. uiowa.edu/~csl/publications/pdf/leesee04.pdf

2004
[19]

Raymond Lister et al. 2004. A multi-national study of reading and tracing skills in novice programmers.SIGCSE Bulletin36, 4 (6 2004), 119–150. https://opus.lib. uts.edu.au/bitstream/10453/4126/3/2004000904.pdf

2004
[20]

Rose Luckin, Wayne Holmes, Marc Griffiths, and Laurie Forcier. 2016. Intelligence Unleashed: An Argument for AI in Education. https://oro.open.ac.uk/50104/

2016
[21]

Tanya Nazaretsky, Mutlu Cukurova, and Giora Alexandron. 2022. An instrument for measuring teachers’ trust in AI-based educational technology. InLAK22: 12th international learning analytics and knowledge conference. 56–66

2022
[22]

Justin Reich. 2020. Failure to Disrupt: Why Technology Alone Can’t Transform Education.Harvard University Press(2020)

2020
[23]

Anthony Robins. 2010. Learning edge momentum: A new account of outcomes in CS1.Computer Science Education20, 1 (3 2010), 37–71

2010
[24]

Jeremy Roschelle, Roy Pea, Christopher Hoadley, Douglas Gordin, and Barbara Means. 2000. Changing How and What Children Learn in School With Computer- Based Technologies.The Future of Children10, 2 (2000), 76–101. https://telearn. hal.science/hal-00190610/document

2000
[25]

1990.The Evolution of American Educational Technology

Paul Saettler. 1990.The Evolution of American Educational Technology. Information Age Publishing

1990
[26]

Winkler and M

R. Winkler and M. Söllner. 2020. Unleashing the potential of chatbots in education: A TAM-based analysis. InProceedings of ICIS

2020
[27]

Yong Xu and Ming Li. 2022. AI in the Classroom: Challenges and Opportunities. Educational Technology Research and Development70 (2022), 123–140

2022
[28]

Ming Yin, Jennifer Wortman Vaughan, and Hanna Wallach. 2019. Understanding the effect of accuracy on trust in machine learning models. InProceedings of the 2019 CHI conference on human factors in computing systems, Stephen A. Brewster, Geraldine Fitzpatrick, Anna L. Cox, and V...

2019
[29]

Ying Zhu and Philip J. Guo. 2021. How AI Tutors Shape Student Reasoning in SE. InACM SIGCSE

2021
[2025]

doi:10.1038/s41598-025-97652-6

AI tutoring outperforms in-class active learning: an RCT introducing a novel research-based design in an authentic educational setting.Scientific Reports 15, 1 (June 2025). doi:10.1038/s41598-025-97652-6

2025 doi

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [1]

2013.International handbook of metacognition and learning technologies

Roger Azevedo and Vincent Aleven. 2013.International handbook of metacognition and learning technologies. Springer. 1000 pages. doi:10.1007/978-1-4419-5546-3

work page doi:10.1007/978-1-4419-5546-3 2013

[2] [2]

Narges Bakhtiar and Sidney K. D’Mello. 2022. A Framework for Trust in Educa- tional AI.International Journal of Artificial Intelligence in Education32 (2022), 1–23

2022

[3] [3]

Patrick Bassner, Eduard Frankford, and Stephan Krusche. 2024. Iris: An AI- Driven Virtual Tutor for Computer Science Education. InProceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1(Milan, Italy)(ITiCSE 2024). Association for Computing Machinery, New York, NY, USA, 394–400. doi:10.1145/3649217.3653543

arXiv 2024

[4] [4]

Michelene TH Chi, Marguerite Roy, and Robert G Hausmann. 2008. Observing tutorial dialogues collaboratively: Insights about human tutoring effectiveness from vicarious learning.Cognitive science32, 2 (3 2008), 301–341

2008

[5] [5]

2001.Oversold and Underused: Computers in the Classroom

Larry Cuban. 2001.Oversold and Underused: Computers in the Classroom. Harvard University Press

2001

[6] [6]

Wei Dai, Jionghao Lin, Hua Jin, Tongguang Li, Yi-Shan Tsai, Dragan Gašević, and Guanliang Chen. 2023. Can Large Language Models Provide Feedback to Students? A Case Study on ChatGPT. In2023 IEEE International Conference on Advanced Learning Technologies (ICALT). 323–325. doi:10.1109/ICALT58122.2023. 00100

arXiv 2023

[7] [7]

Fred D. Davis. 1989. Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology.MIS Quarterly13, 3 (9 1989), 319–340. doi:10.2307/249008 Publisher: Management Information Systems Research Center, University of Minnesota

doi:10.2307/249008 1989

[8] [8]

Eduard Frankford, Clemens Sauerwein, Patrick Bassner, Stephan Krusche, and Ruth Breu. 2024. AI-Tutoring in Software Engineering Education, In Proceedings of the 46th International Conference on Software Engineering: Software Engineer- ing Education and Training (Lisbon Portugal, 2024-04-14).2024 IEEE/ACM 46th International Conference on Software Engineeri...

arXiv 2024