Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

MOSAIC-F: A Framework for Enhancing Students' Oral Presentation Skills through Personalized Feedback

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A four-step feedback loop blends rubrics, sensor data, and AI to personalize oral-presentation coaching.

desk verdict Clear framework write-up, but the effectiveness claim is unsupported: the paper's own text defers the analysis to future work. read the letter →

arxiv 2506.08634 v1 pith:WDJLD6ER submitted 2025-06-10 cs.HC cs.AIcs.CV

Alvaro Becerra , Daniel Andres , Pablo Villegas , Roberto Daza , Ruth Cobos This is my paper
classification cs.HCcs.AIcs.CV
keywords multimodallearninganalyticsoralpresentationfeedbackpeerassessmentself-assessmentgenerativeAIwearablesensorseyetrackingpersonalized
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

MOSAIC-F is a proposed workflow for giving students personalized feedback on skills such as oral presentation. It combines human judgment, with peers and professors scoring a standardized rubric, with multimodal data collected during the performance, including video, audio, gaze, heart rate, motion, and interaction logs. An AI layer synthesizes these sources into written feedback covering strengths, weaknesses, and an action plan, and students then watch their own video, self-assess, and compare their scores with peers, professors, and class averages. The paper argues that this combination makes feedback more accurate, personalized, and actionable than traditional evaluation alone. It reports the design and a feasibility case study with 46 students, with the multimodal analyses still planned rather than completed.

What carries the argument

The load-bearing mechanism is the four-step MOSAIC-F workflow itself. Step 1 uses the AICoFe web system to collect quantitative and qualitative rubric ratings from peers and professors. Step 2 captures synchronized multimodal signals, including frontal and room video, ambient audio, smartwatch heart rate and motion data, eye tracking, and keyboard and click logs. Step 3 uses the GePeTo generative-AI module to turn ratings and observations into structured feedback, with strengths, improvement areas, and an action plan, while professors retain oversight of the automatically generated text. Step 4 presents the student with their recording, their own rubric-based self-assessment, and comparative visualizations against peers, professors, and class averages. The data-based analyses that would connect Step 2 to Step 3, such as head-pose attention, posture, audio features, transcription patterns, heart-rate variation, gaze, and interaction logs, are explicitly described as planned analyses.

What would settle it

Take two groups of students giving comparable presentations, give one group MOSAIC-F feedback and the other rubric-only feedback, and compare how actionable and useful the students find the comments; if the groups do not differ, or if the sensor-based measures of attention and stress do not match independent human ratings of the same recordings, the framework's core claim is not supported. A simpler check is whether the planned heart-rate comparisons actually separate known stressful segments, such as audience questions, from calmer ones.

Watch

Extended reading notes

Core claim

The central claim is that feedback improves when it is generated through a four-step loop: standardized rubric assessments by peers and professors; synchronized collection of multimodal and physiological data during the activity; AI-generated feedback that merges the human scores with data-derived insights such as posture, speech patterns, stress, and cognitive load; and a self-assessment phase where students watch their recorded performance and compare their own evaluation with external scores and class averages. The authors state that this combination of human-based and data-based evaluation enables more accurate, personalized, and actionable feedback. The paper's evidence at this stage is a feasibility case study with 46 engineering students; the planned analyses of the sensor data are described, but their results are not yet reported.

Load-bearing premise

Everything depends on whether raw signals like heart rate, posture, gaze, and speech can be read reliably as evidence of stress, attention, and cognitive load during a presentation, and the paper says the analyses that would show this are planned, not yet completed.

Editorial extensions

If this is right

  • If MOSAIC-F works as intended, students receive feedback that combines a human rater's judgment with objective behavioral signals, reducing reliance on any single evaluator's subjective impression.
  • AI-generated feedback can be produced at scale and then checked by professors, making personalized, rubric-aligned comments feasible for larger classes.
  • Video review plus comparative visualizations gives students a structured way to reconcile their self-assessment with external scores and class averages, which should support self-regulated improvement.
  • The same four-step loop could be transferred to other competency areas such as teamwork, which the paper itself names as future work.
  • The planned statistical tests on heart rate across presentation phases, if carried out, would provide an evidence base for claims about stress and engagement at specific moments.

Reading between the lines

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

  • [Editorial inference] The framework's actual value will hinge on whether sensor-derived constructs such as attention, stress, and cognitive load can be validated against independent measures; a direct test would correlate head-pose attention estimates with the observer's gaze annotations or with self-report.
  • [Editorial inference] Because the feedback is drafted by a language model and then reviewed by professors, the design implicitly treats human oversight as a guard against hallucinated recommendations; comparing reviewed and unreviewed feedback would test whether that oversight changes student trust or usefulness ratings.
  • [Editorial inference] The eye-tracking component measures only one observer's gaze, so its attention findings should be read as a proof of concept rather than a population-level measure of audience engagement.
  • [Editorial inference] If the physiological mapping succeeds, the same sensor stack could plausibly be reused outside presentations, for example in interview coaching or team-collaboration assessment, but the paper does not yet claim this.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The manuscript introduces MOSAIC-F, a four-step multimodal feedback framework intended to improve students' oral presentation skills through peer and professor rubric assessments, multimodal sensor data collection, AI-generated personalized feedback, and self-assessment with visualization. The paper describes the framework, its application in a case study with 46 engineering students, and a plan of multimodal analyses (head pose, posture, audio, heart rate, gaze, interaction logs, slides). The authors claim that combining human-based and data-based evaluation enables more accurate, personalized, and actionable feedback, and state that they tested the framework in this oral-presentation context.

Significance. If the claimed effectiveness were demonstrated, the framework would be a valuable contribution to multimodal learning analytics and automated feedback research. The paper's related-work coverage is broad and organized, and the framework's explicit four-step workflow with human oversight of AI feedback is a useful design contribution. The strengths are the clear description of the sensor setup, the ethical data-collection protocol, and the honest acknowledgment that the in-depth multimodal analyses and effectiveness assessment are future work. However, the central claim—that the framework enables more accurate, personalized, and actionable feedback—is asserted rather than demonstrated anywhere in the manuscript. No outcome data, no comparison condition, no validity evidence, and no completed multimodal analysis are reported. The paper therefore does not currently provide scientific evidence for its main claim.

major comments (4)
  1. [Abstract, Section 1, Section 5] The abstract and Section 1 assert that MOSAIC-F "enables more accurate, personalized and actionable feedback," but the manuscript reports no empirical results supporting this assertion. Section 4 explicitly states "we are planning to conduct the following analyses" for every multimodal channel, and Section 5 states that "as part of our future work, we will conduct an in-depth analysis of the multimodal data collected during the case study to assess the effectiveness and accuracy of the feedback mechanisms." Thus, the paper's own text contradicts the effectiveness claim; no evidence of accuracy, personalization, or actionability is presented.
  2. [Section 3.3, Section 4] Section 3.3 describes a data-based feedback report generated from MMLA, but Section 4 lists the analyses as planned rather than completed. The described feedback pipeline in Section 3.3 uses only AICoFe rubric input and GePeTo's generative AI to produce text based on human quantitative and qualitative evaluations. There is no indication that any sensor-derived insight (posture, speech, stress, cognitive load) was actually extracted, mapped to a pedagogical construct, or included in the feedback students received. The central claimed benefit of combining human- and data-based evaluation is therefore unsupported by the implementation described.
  3. [Section 1, Section 5] The paper states "We tested MOSAIC-F" and reports a case study with 46 students, but the only outcome reported is that the implementation "allowed us to validate the framework's feasibility." Feasibility is a much weaker claim than the effectiveness claim in the abstract. No measure of learning gains, feedback quality, student perception, or comparison with a baseline feedback condition is provided. As a journal submission, the absence of any evaluation outcome leaves the framework's value unsubstantiated.
  4. [Section 3.3, Section 2.3] The framework relies on the authors' own AICoFe rubric system and GePeTo LLM tool as the core feedback generation components. The paper does not provide validity evidence for the rubric (e.g., inter-rater reliability) or for GePeTo's generated feedback (e.g., alignment with expert feedback, consistency, or educational impact). Given that Section 2.3 itself reviews evidence that students perceive AI feedback as less credible and trustworthy, the paper should report at least a basic evaluation of the feedback quality generated by these tools in the case study.
minor comments (5)
  1. [Section 3] The sentence "MOSAIC-F use a four step workflow" should be "MOSAIC-F uses a four-step workflow."
  2. [References] Reference [2] lists "S. Askew" twice; this appears to be a formatting error. Several other references (e.g., [20], [22]) have minor punctuation inconsistencies that should be cleaned up.
  3. [Section 3.2] The roles and sensor descriptions are clear, but the relationship between the "external observer" with the Tobii glasses and the "observer" (research assistant) who annotates events is confusing; clarify whether these are the same person or two different roles.
  4. [Section 4] The phrase "we are planning to conduct the following analyses" is inconsistent with the surrounding text, which describes these analyses in the present/future tense as if they are part of the case study. Because these are planned, the section should be labeled as an analysis plan or intended analyses, not as results.
  5. [Section 3.4] The phrase "students are invited to reflect on the feedback received and indicate whether they agree with the assessment" is an interesting design element, but no data from this reflection step is reported; if the case study is claimed as a test, this outcome should be included.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper's effectiveness claim is an unsubstantiated assertion, not a reduction of outputs to inputs.

full rationale

The paper's central benefit claim ('By combining human-based and data-based evaluation techniques, this framework enables more accurate, personalized and actionable feedback') is asserted in the abstract but not derived from any measured quantity in the manuscript. The data-based component is explicitly pending: Section 4 states 'we are planning to conduct the following analyses', and Section 5 says 'we will conduct an in-depth analysis of the multimodal data collected during the case study to assess the effectiveness and accuracy of the feedback mechanisms integrated in MOSAIC-F'. There is no fitted parameter renamed as a prediction, no equation that equates an output to an input, and no uniqueness theorem imported from the authors' prior work. The self-citations to AICoFe [36] and GePeTo [37] name the implementation components used in the workflow; they are not used to prove the framework's effectiveness, and the paper even labels the case-study outcome as feasibility ('This initial implementation allowed us to validate the framework's feasibility'). The gap between the abstract's language and the paper's own future-work statements is a correctness/evidence concern, not a circularity concern, because the claimed benefit is not shown to be equivalent to its inputs by construction. Accordingly, no specific circular step can be exhibited, and the appropriate circularity finding is negative.

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

The central claim rests on assumed validity of multimodal-to-construct mappings, the pedagogical soundness of the fine-tuned LLM, the rubric's validity, and the benefit of video self-assessment. None of these are established with data in this paper.

assumptions (4)
  • domain assumption Multimodal signals can be mapped to attention, stress, and cognitive load.
    Section 4 describes planned analyses that interpret head pose, heart rate, and gaze as these constructs without calibration or validation.
  • domain assumption The fine-tuned ChatGPT model generates pedagogically sound feedback.
    Section 3.3 asserts the model is fine-tuned so outputs are pedagogically sound, but no evaluation of the generated feedback is provided beyond later professor review.
  • domain assumption The AICoFe rubric is a valid measure of oral presentation skill.
    Section 3.1 uses the rubric as the shared instrument for peer, professor, and self-assessment, but the paper offers no psychometric validation or inter-rater reliability data.
  • domain assumption Video-based self-assessment improves learning and self-awareness.
    Section 3.4 relies on reference [4] for this effect rather than measuring it within the reported case study.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MOSAIC-F: A Framework for Enhancing Students' Oral Presentation Skills through Personalized Feedback." pith.science (2026). https://pith.science/paper/WDJLD6ER

@misc{pith2026250608634,
  author       = {Pith},
  title        = {Pith review of: MOSAIC-F: A Framework for Enhancing Students' Oral Presentation Skills through Personalized Feedback},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WDJLD6ER}},
  note         = {Machine review of arXiv:2506.08634}
}
read the original abstract

In this article, we present a novel multimodal feedback framework called MOSAIC-F, an acronym for a data-driven Framework that integrates Multimodal Learning Analytics (MMLA), Observations, Sensors, Artificial Intelligence (AI), and Collaborative assessments for generating personalized feedback on student learning activities. This framework consists of four key steps. First, peers and professors' assessments are conducted through standardized rubrics (that include both quantitative and qualitative evaluations). Second, multimodal data are collected during learning activities, including video recordings, audio capture, gaze tracking, physiological signals (heart rate, motion data), and behavioral interactions. Third, personalized feedback is generated using AI, synthesizing human-based evaluations and data-based multimodal insights such as posture, speech patterns, stress levels, and cognitive load, among others. Finally, students review their own performance through video recordings and engage in self-assessment and feedback visualization, comparing their own evaluations with peers and professors' assessments, class averages, and AI-generated recommendations. By combining human-based and data-based evaluation techniques, this framework enables more accurate, personalized and actionable feedback. We tested MOSAIC-F in the context of improving oral presentation skills.

Figures

Figures reproduced from arXiv: 2506.08634 by the authors.

Figure 1
Figure 1. MOSAIC-F workflow diagram. The rubric includes different items, such as eye contact, attention capture or clarity of the opening, and each item is assessed using a 5-point Likert scale, with predefined descriptions for each level to ensure consistency among all the evaluators. Additionally, peers and professors should provide qualitative observations for each item. 3.2. Multimodal Data Collection MOSAIC-F uses sever… view at source ↗
Figure 2
Figure 2. Acquisition setup illustrating the different roles and sensors used. GePeTo is built on a fine-tuned version of the ChatGPT language model, specifically adapted for oral presentations. The model has been fine-tuned using feedback examples, ensuring that the generated outputs are pedagogically sound, contextually relevant, and aligned with the evaluation rubric used in AICoFe. In addition to this human-based feedback… view at source ↗

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. AI-based Multimodal Biometrics for Detecting Smartphone Distractions: Application to Online Learning

    cs.CY 2025-06 conditional novelty 5.0 of 10

    On 66 learners, a multimodal model of EEG, heart rate and head pose detects instructed phone use during online learning with 91% accuracy, versus 87% for head pose alone.

Reference graph

Works this paper leans on

37 extracted references · 35 canonical work pages · cited by 1 Pith paper

  1. [1]

    Hattie, H

    J. Hattie, H. Timperley, The Power of Feedback, Review of Educational Research 77 (2007) 81–112

    work page 2007

  2. [2]

    Askew, S

    S. Askew, S. Askew, Feedback for Learning, Technical Report, RoutledgeFalmer London, 2000

    work page 2000

  3. [3]

    Henderson, T

    M. Henderson, T. Ryan, M. Phillips, The Challenges of Feedback in Higher Education, Assessment & Evaluation in Higher Education (2019)

    work page 2019

  4. [4]

    Tailab, N

    M. Tailab, N. Marsh, Use of Self-Assessment of Video Recording to Raise Students’ Awareness of Development of Their Oral Presentation Skills, Higher Education Studies 10 (2020)

    work page 2020

  5. [5]

    Giannakos, D

    M. Giannakos, D. Spikol, D. Di Mitri, K. Sharma, X. Ochoa, R. Hammad, The Multimodal Learning Analytics Handbook, Springer, 2022

    work page 2022

  6. [6]

    C. Lang, G. Siemens, A. Wise, D. Gasevic, Handbook of Learning Analytics (2017)

    work page 2017

  7. [7]

    Baró-Solé, A

    X. Baró-Solé, A. E. Guerrero-Roldan, J. Prieto-Blázquez, A. Rozeva, O. Marinov, C. Kiennert, P.-O. Rocher, J. Garcia-Alfaro, Integration of an Adaptive Trust-Based E-Assessment System into Virtual Learning Environments—The TeSLA Project Experience, Internet Technology Letters 1 (2018) e56

    work page 2018

  8. [8]

    R. Daza, A. Morales, R. Tolosana, L. F. Gomez, J. Fierrez, J. Ortega-Garcia, edBB-Demo: Biometrics and Behavior Analysis for Online Educational Platforms, in: Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 2023, pp. 16422–16424

    work page 2023

Show all 37 references
  1. [9]

    Becerra, R

    A. Becerra, R. Daza, R. Cobos, A. Morales, M. Cukurova, J. Fierrez, M2LADS: A System for Generating Multimodal Learning Analytics Dashboards, in: 2023 IEEE 47th Annual Computers, Software, and Applications Conference (COMPSAC), IEEE, 2023, pp. 1564–1569

    2023

  2. [10]

    Becerra, R

    A. Becerra, R. Daza, R. Cobos, A. Morales, J. Fierrez, M2LADS Demo: A System for Generating Multimodal Learning Analytics Dashboards, arXiv preprint arXiv:2502.15363 (2025)

    2025 arXiv

  3. [11]

    Becerra, R

    A. Becerra, R. Daza, R. Cobos, A. Morales, J. Fierrez, User Experience Study Using a System for Generating Multimodal Learning Analytics Dashboards, in: Proceedings of the XXIII International Conference on Human Computer Interaction, 2023, pp. 1–2

    2023

  4. [12]

    Becerra, J

    A. Becerra, J. Irigoyen, R. Daza, R. Cobos, A. Morales, J. Fierrez, M. Cukurova, Biometrics and Behavioral Modelling for Detecting Distractions in Online Learning, in: Proc. Simposio Internacional de Informática Educativa (SIIE), VII Congreso Español de Informática, 2024

    2024

  5. [13]

    R. Daza, A. Becerra, R. Cobos, J. Fierrez, A. Morales, IMPROVE: Impact of Mobile Phones on Remote Online Virtual Education, arXiv preprint arXiv:2412.14195 (2024)

    2024 arXiv

  6. [14]

    Navarro, A

    M. Navarro, A. Becerra, R. Daza, R. Cobos, A. Morales, J. Fierrez, VAAD: Visual Attention Analysis Dashboard Applied to E-Learning, in: 2024 International Symposium on Computers in Education (SIIE), IEEE, 2024, pp. 1–6

    2024

  7. [15]

    Spikol, E

    D. Spikol, E. Ruffaldi, G. Dabisias, M. Cukurova, Supervised Machine Learning in Multimodal Learning Analytics for Estimating Success in Project-Based Learning, Journal of Computer Assisted Learning 34 (2018) 366–377

    2018

  8. [16]

    F. P. García, O. Cánovas, F. J. G. Clemente, Exploring AI Techniques for Generalizable Teaching Practice Identification, IEEE Access (2024)

    2024

  9. [17]

    Bosch, Y

    N. Bosch, Y. Chen, S. D’Mello, It’s Written on Your Face: Detecting Affective States from Facial Expressions While Learning Computer Programming, in: Intelligent Tutoring Systems: 12th International Conference, ITS 2014, Honolulu, HI, USA, June 5-9, 2014. Proceedings 12, Sprin...

    2014

  10. [18]

    C. C. Ekin, O. F. Cantekin, E. Polat, S. Hopcan, Artificial Intelligence in Education: A Text Mining-Based Review of the Past 56 Years, Education and Information Technologies (2025) 1–43

    2025

  11. [19]

    M. Kim, S. Kim, S. Lee, Y. Yoon, J. Myung, H. Yoo, H. Lim, J. Han, Y. Kim, S.-Y. Ahn, et al., LLM- Driven Learning Analytics Dashboard for Teachers in EFL Writing Education, arXiv preprint arXiv:2410.15025 (2024)

    2024 arXiv

  12. [20]

    Schneider, D

    J. Schneider, D. Börner, P. Van Rosmalen, M. Specht, Presentation Trainer: What Experts and Computers Can Tell About Your Nonverbal Communication, Journal of Computer Assisted Learning 33 (2017) 164–177

    2017

  13. [21]

    Di Mitri, J

    D. Di Mitri, J. Schneider, N. Mouhammad, S. Hummel, M. Alomari, M. A. Ali, M. H. R. Masum, H. Arif, M. Rose, R. Klemke, Enhance Your Presentation Skills with Presentable, in: Proceedings of the 15th International Conference on Learning Analytics & Knowledge (LAK 2025), Demo Tr...

    2025

  14. [22]

    Ochoa, F

    X. Ochoa, F. Domínguez, B. Guamán, R. Maya, G. Falcones, J. Castells, The RAP System: Automatic Feedback of Oral Presentation Skills Using Multimodal Analysis and Low-Cost Sensors, in: Proceedings of the 8th International Conference on Learning Analytics and Knowledge, 2018, p...

    2018

  15. [23]

    Ochoa, H

    X. Ochoa, H. Zhao, OpenOPAF: An Open-Source Multimodal System for Automated Feedback for Oral Presentations, Journal of Learning Analytics 11 (2024) 224–248

    2024

  16. [24]

    R. Daza, L. Shengkai, A. Morales, J. Fierrez, K. Nagao, SMARTe-VR: Student Monitoring and Adaptive Response Technology for e-learning in Virtual Reality, in: Proc. AAAI Workshop on Artificial Intelligence for Education, 2025

    2025

  17. [25]

    Yokoyama, K

    Y. Yokoyama, K. Nagao, VR Presentation Training System Using Machine Learning Techniques for Automatic Evaluation, International Journal of Virtual and Augmented Reality (IJVAR) (2021)

    2021

  18. [26]

    VanLehn, The Relative Effectiveness of Human Tutoring, Intelligent Tutoring Systems, and Other Tutoring Systems, Educational Psychologist 46 (2011) 197–221

    K. VanLehn, The Relative Effectiveness of Human Tutoring, Intelligent Tutoring Systems, and Other Tutoring Systems, Educational Psychologist 46 (2011) 197–221

    2011

  19. [27]

    Zhang, D

    H. Zhang, D. Litman, Co-Attention Based Neural Network for Source-Dependent Essay Scoring, arXiv preprint arXiv:1908.01993 (2019)

    2019 arXiv

  20. [28]

    Rüdian, J

    S. Rüdian, J. Podelo, J. Kužílek, N. Pinkwart, Feedback on Feedback: Student’s Perceptions for Feedback from Teachers and Few-Shot LLMs, in: Proceedings of the 15th International Learning Analytics and Knowledge Conference, 2025, pp. 82–92

    2025

  21. [29]

    Nazaretsky, P

    T. Nazaretsky, P. Mejia-Domenzain, V. Swamy, J. Frej, T. Käser, AI or Human? Evaluating Student Feedback Perceptions in Higher Education, in: European Conference on Technology Enhanced Learning, Springer, 2024, pp. 284–298

    2024

  22. [30]

    Nazaretsky, P

    T. Nazaretsky, P. Mejia-Domenzain, V. Swamy, J. Frej, T. Käser, The Critical Role of Trust in Adopting AI-Powered Educational Technology for Learning: An Instrument for Measuring Student Perceptions, Computers and Education: Artificial Intelligence (2025) 100368

    2025

  23. [31]

    Steiss, T

    J. Steiss, T. Tate, S. Graham, J. Cruz, M. Hebert, J. Wang, Y. Moon, W. Tseng, M. Warschauer, C. B. Olson, Comparing the Quality of Human and ChatGPT Feedback of Students’ Writing, Learning and Instruction 91 (2024) 101894

    2024

  24. [32]

    T. Wan, Z. Chen, Exploring Generative AI Assisted Feedback Writing for Students’ Written Responses to a Physics Conceptual Question with Prompt Engineering and Few-Shot Learning, Physical Review Physics Education Research 20 (2024) 010152

    2024

  25. [33]

    C. D. Kloos, C. Alario-Hoyos, I. Estévez-Ayres, P. Callejo-Pinardo, M. A. Hombrados-Herrera, P. J. Muñoz Merino, P. M. Moreno-Marcos, M. Muñoz Organero, M. B. Ibáñez, How Can Generative AI Support Education?, in: 2024 IEEE Global Engineering Education Conference (EDUCON), IEEE...

    2024

  26. [34]

    Ogata, C

    H. Ogata, C. Liang, Y. Toyokawa, C.-Y. Hsu, K. Nakamura, T. Yamauchi, B. Flanagan, Y. Dai, K. Takami, I. Horikoshi, et al., Co-Designing Data-Driven Educational Technology and Practice: Reflections from the Japanese Context, Technology, Knowledge and Learning 29 (2024) 1711–1732

    2024

  27. [35]

    Topali, A

    P. Topali, A. Ortega-Arranz, Y. Dimitriadis, S. Villagrá-Sobrino, A. Martínez-Monés, J. I. Asensio- Pérez, Unlock the feedback potential: Scaling effective teacher-led interventions in massive educational contexts, in: Innovating Assessment and Feedback Design in Teacher Educa...

    2023

  28. [36]

    Becerra, R

    A. Becerra, R. Cobos, Enhancing the Professional Development of Engineering Students Through an AI-Based Collaborative Feedback System, in: 2025 IEEE Global Engineering Education Conference (EDUCON), IEEE, 2025, pp. 1–9

    2025

  29. [37]

    Becerra, Z

    A. Becerra, Z. Mohseni, J. Sanz, R. Cobos, A Generative AI-Based Personalized Guidance Tool for Enhancing the Feedback to MOOC Learners, in: 2024 IEEE Global Engineering Education Conference (EDUCON), IEEE, 2024, pp. 1–8

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.