Pith. sign in

REVIEW 3 major objections 5 minor 3 cited by

LLM-powered Multi-agent Framework for Goal-oriented Learning in Intelligent Tutoring System

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read GenMentor, a multi-agent LLM framework for goal-oriented learning, maps goals to skills, profiles learners, and adapts paths and content via a learner simulator, outperforming prompting and RAG baselines.

desk verdict A coherent, well-engineered ITS system paper whose central evaluation is a closed GPT-4o loop; useful as a design reference, not as a proof of effectiveness. read the letter →

arxiv 2501.15749 v1 pith:QZ6VEBNZ submitted 2025-01-27 cs.AI cs.MA

Tianfu Wang , Yi Zhan , Jianxun Lian , Zhengyu Hu , Nicholas Jing Yuan , Qi Zhang , Xing Xie , Hui Xiong This is my paper · ORCID
classification cs.AIcs.MA
keywords IntelligentTutoringSystemGoal-orientedLearningLargeLanguageModelMulti-agentSkillGapIdentificationPathSchedulingPersonalizedContentGenerationLearnerModeling
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 proposes GenMentor, an LLM-powered multi-agent framework that tries to deliver goal-oriented, personalized learning inside intelligent tutoring systems. The central idea is to move from reactive chatbots and static curricula to proactive guidance: map the learner's goal to the required skills, identify the gap against the learner's profile, schedule an adaptive learning path, and generate tailored content. GenMentor's skill identifier is a fine-tuned LLM trained on a custom goal-to-skill dataset built from job postings with chain-of-thought reasoning tracks. The paper reports that this design outperforms direct prompting, CoT prompting, and retrieval-augmented baselines on automated and human evaluations, and that a deployed version with professional learners is rated as providing clearer guidance and more targeted resources.

What carries the argument

The load-bearing mechanism is the iterative adaptation loop driven by the learner simulator, an LLM agent that role-plays the learner using the current profile $U_t$ and anticipates feedback on the scheduled path and drafted content. This simulated feedback lets the path scheduler evolve the learning path $\mathbf{L}_t$ and the content creator refine each document before a real learner sees it, without requiring direct user input at every step. A second load-bearing piece is the CoT-fine-tuned skill identifier, which maps an abstract goal $G$ to a skill set $S'$ using reasoning tracks derived from job-posting data, and then to the skill gap $\Delta S_0$ by subtracting already-mastered competencies $S_0$.

What would settle it

A randomized controlled trial in which professionals with the same learning goal are assigned to GenMentor or the strongest baseline, with pre- and post-tests of the target skills, would settle the central claim: if the GenMentor group shows no significantly larger skill gains despite higher satisfaction, the claimed effectiveness would not hold. A second check is to collect real learner feedback on the same path and content items and measure its correlation with the learner simulator's predictions; low correlation would undercut the adaptation loop.

Watch

Extended reading notes

Core claim

GenMentor claims to close the skill gap $\Delta S_0 = S' - S_0$ efficiently by orchestrating four LLM agents: a skill identifier that maps the goal to a complete, aligned skill set via CoT fine-tuning; a learner profiler that tracks cognitive status, preferences, and behavioral patterns; a path scheduler that refines the learning path with feedback from an LLM learner simulator; and a content creator that explores goal-related knowledge, drafts sections with retrieval augmentation, and integrates them with simulated-learner refinement. The paper's key result is that this pipeline, evaluated by GPT-4o as an automated judge and by human raters, achieves higher recall and precision in goal-to-skill mapping, higher progression and engagement scores for learning paths, and higher content quality and personalization scores than direct prompting, CoT prompting, and RAG-based baselines. End-to-end, a post-deployment human study with twenty professional learners rated the system highly on goal alignment, learning efficiency, and user experience.

Load-bearing premise

The framework's reported superiority rests on three measurement assumptions: that GPT-4o's automated scores reflect what human judges would say (only five of seven metrics passed the correlation check), that the learner simulator's mimicked feedback matches real learners' reactions, and that self-reported satisfaction in the twenty-person study is a valid proxy for actual learning effectiveness.

Editorial extensions

If this is right

  • With a faithful learner simulator, the system can personalize paths and content with minimal live feedback, reducing the burden on the learner.
  • The CoT-fine-tuned goal-to-skill mapping could be reused across occupations, since it is trained on job-posting summaries and transfers to new goals without retraining the whole pipeline.
  • The exploration-drafting-integration content loop, combined with retrieval, produces documents that are more goal-aligned and personalized than outline-based or RAG-only generation.
  • Deployment in a product and an app shows the framework is feasible in professional settings, not just in offline evaluations.

Reading between the lines

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

  • Editorial: If the learner simulator's feedback truly matches real learner reactions, the same role-play-amplified adaptation loop could be applied to other sparse-feedback domains such as onboarding, career pivots, or compliance training.
  • Editorial: The goal-to-skill dataset recipe—harvesting job postings, extracting summaries and skills, and inserting CoT tracks—is a transferable method for building intent-to-competency maps in fields beyond education, for example talent matching or internal mobility.
  • Editorial: The paper's own validation shows only 5 of 7 automated metrics correlate significantly with human grading; a stronger test would be a randomized controlled trial measuring actual skill acquisition on pre/post tests rather than satisfaction.
  • Editorial: A direct comparison of the learner simulator against real learner feedback on the same materials would quantify the simulator's fidelity and tell whether the adaptive loop overfits to its own simulation.
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

3 major / 5 minor

Summary. The paper presents GenMentor, an LLM-powered multi-agent framework for goal-oriented learning in intelligent tutoring systems. The system maps a learner's goal to required skills via a fine-tuned LLM trained on a custom goal-to-skill dataset, identifies the learner's skill gap, maintains an adaptive learner profile, schedules an evolvable learning path using feedback from an LLM-based learner simulator, and generates personalized content through an exploration-drafting-integration mechanism with RAG. The authors report automated evaluations with GPT-4o as judge, a small human validation of the automated scores, a five-participant pairwise preference study, and an end-to-end study with 20 professional learners, along with a practical deployment in a commercial product.

Significance. If the effectiveness claims were established, GenMentor would be a useful systems contribution to the WWW community: it addresses a real need for goal-oriented professional learning, integrates several LLM agents into a coherent pipeline, and includes a concrete deployment. The framework's design choices, such as CoT-based fine-tuning for goal-to-skill mapping and simulator-driven path/content refinement, are reasonable and worth studying. However, the current evidence base is not strong enough to support the central claim of superiority over baselines: the goal-to-skill evaluation is circular, the human validation is thin and covers only a subset of metrics, the learner simulator is unvalidated, and the end-to-end human study is a small self-report study without a control or objective learning measure. These are load-bearing gaps, not presentational issues.

major comments (3)
  1. [§5.2.2 and §A.1] The evaluation of goal-to-skill mapping is circular. The training data for the fine-tuned skill identifier is created by using GPT-4o to extract job summaries and skill lists from LinkedIn postings (§A.1), and the held-out test ground truth in §5.2.2 is again GPT-4o-extracted skill requirements, with GPT-4o used as the judge for Recall, Precision, and Goal Alignment. Consequently, the reported superiority in Table 1 may only demonstrate that GenMentor reproduces GPT-4o's labeling style rather than that it identifies skills that are correct for real job goals. This is load-bearing because skill-gap identification is the first stage of the pipeline and its errors propagate into path scheduling and content generation. The authors should provide an independent human-annotated gold standard, or at minimum human evaluation of all three metrics, before claiming that GenMentor improves goal-to-skill mapping.
  2. [§5.2.5 and §A.2] The human validation of the automated evaluation is too weak to support the paper's quantitative claims. It uses only 20 Python-developer samples and two raters, and it validates only Goal Alignment for skill mapping, not Recall or Precision. Even for the metrics that are validated, Pearson correlations range from 0.38 to 0.52 with only 5 of 7 metrics significant, which does not establish that GPT-4o scores agree with human judgment across the occupations and output types used in Tables 1-2 and Figure 4. In addition, the automated scores in Tables 1-2 and Figure 4 are reported without confidence intervals, standard deviations, or significance tests, so differences such as 4.56 vs. 4.38 for GPT-4o Progression in Table 2 may not be statistically reliable. The authors should report distributional statistics and formal comparisons, and they should validate the automated scores on all metrics and occupations used in the evaluation.
  3. [§4.3.1 and §6.2-6.3] The learner simulator is a core component of the claimed improvements, but the paper provides no evidence that the simulated feedback matches real learner reactions. The ablation comparisons in §5.2.3-5.2.4 only show that including mimicked feedback changes the output, not that it improves learning outcomes. The end-to-end human study in §6.2-6.3 does not fill this gap: it has 20 participants, no control condition, no pre/post knowledge test, and relies entirely on self-reported satisfaction and perceived efficiency. Statements such as 'more than 80% noting enhanced efficiency' (§6.3) are therefore not supported as objective learning gains. The authors should either add a controlled comparison or an objective learning measure, or substantially weaken the causal claims and explicitly acknowledge the simulator-validation limitation.
minor comments (5)
  1. [§5.2.2] The sentence 'we include a variant of GenMentor, (3) w/o Tracks, which removes track-based guidance to assess its impact. All methods use the same learner information and learning goal as input. To evaluate the quality of goal-to-skill mapping...' contains a garbled phrase: 'CoTPrompt, where the CoT reasoning is integrated into the prompt no or yes.' This should be rewritten for clarity.
  2. [Table 3 in §A.2] The p-values are formatted as '< 4−2', '< 2−2', etc., which is ambiguous and nonstandard. They should be reported as normal decimal values or in scientific notation, e.g., p < 0.04 or p = 0.037.
  3. [§6.4.1] The interview findings contain duplicated and incomplete parentheticals, e.g., '(highlighted by 15 participants) (noted by 13 participants)' and a fragment 'Participants identified scenarios where GenMentor excels... (noted by 13 participants).' Clean up the prose and ensure each participant count is attributed to a single, clearly specified finding.
  4. [§6.3 and Figure 6] The figure reports means and standard deviations as '4.6 ± 0.8' but does not show the response distribution or the number of participants per item; given the small N=20, a dot plot or histogram of individual responses would be more informative than a bar chart.
  5. [§5.1] The implementation section reports the fine-tuning batch size and maximum epochs but omits learning rate, number of training steps, and data splits for the goal-to-skill dataset. These details are needed for reproducibility, especially since the fine-tuned model is a central component.

Circularity Check

2 steps flagged · score 6.0 of 10

Goal-to-skill evaluation is a closed loop: GPT-4o creates the training labels, GPT-4o extracts the test 'ground truth,' and GPT-4o judges the scores; human validation is too thin to break the loop.

  1. fitted input called prediction [§4.1.1, §5.2.1–5.2.2, Appendix A.1–A.2]
    "we employ GPT4o to extract the job summaries paired with their corresponding skill lists (A.1). ... we regard the skill requirements extracted from job postings as ground truth and use LLM to measure three metrics: Recall ... Precision ... and Goal Alignment (5.2.2). ... Following prior works [12, 25, 38], we use GPT4o as an automated evaluator due to its strong alignment with human judgments (5.2.1)."

    The skill identifier is fine-tuned on GPT-4o-extracted <job summary, skill list> pairs from LinkedIn postings. The evaluation then compares the fine-tuned model's outputs against GPT-4o-extracted skill requirements from held-out postings, with GPT-4o judging Recall, Precision, and Goal Alignment. This measures how well the model reproduces GPT-4o's labeling style, not correctness against an independent gold standard. Human validation (A.2) covers only Goal Alignment on 20 Python-developer samples (Pearson 0.51); Recall and Precision are not human-validated. The central claim of accurate goal-to-skill mapping is therefore partially circular.

  2. other [§4.3.1, §5.2.1, §5.2.3–5.2.4, §6.3]
    "leveraging real-time learner profiles U_t, a learner simulator employs the role-playing method to anticipate learner feedback for delivered resources [24]. This simulation serves as a proxy to optimize delivered resources without requiring direct learner feedback (4.3.1). ... Following prior works [12, 25, 38], we use GPT4o as an automated evaluator due to its strong alignment with human judgments (5.2.1)."

    Learning paths and content are refined using feedback from an LLM-based learner simulator, and then the same LLM family (GPT-4o) is used as the automated judge of Progression, Engagement, Personalization, and related metrics. The reported superiority of GenMentor on these axes thus partly reflects an LLM's simulated learner preferences being scored by an LLM's judgment, rather than observed learning outcomes. The 20-participant human study is self-reported, has no control group, and includes no pre/post knowledge test, so it only partially breaks this closed evaluation loop.

full rationale

GenMentor's architecture is not definitionally circular: the goal-to-skill fine-tuning, adaptive learner profiling, and path/content generation are genuine pipeline components, and no load-bearing self-citation or imported uniqueness theorem is used. The principal circularity lies in the evaluation of the central skill-mapping claim. The model is fine-tuned on GPT-4o-extracted job-summary/skill pairs (A.1) and then tested against GPT-4o-extracted skill requirements from held-out job postings, with GPT-4o as the judge (5.2.1–5.2.2). This establishes agreement with GPT-4o's labeling behavior, not accuracy against an independent gold standard. The human validation in A.2 is limited to 20 Python-developer samples, with statistically significant correlation on only 5 of 7 metrics; for goal-to-skill mapping it validates only Goal Alignment, not Recall or Precision. The path and content evaluations add a second loop: an LLM learner simulator generates the feedback used to refine outputs, and GPT-4o then scores those outputs. The 20-participant human study is self-reported, uncontrolled, and lacks a knowledge test, so it only partially breaks the loop. Overall, the central empirical claim is partially circular because the teacher, the ground truth, and the judge are all the same model family.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical or conceptual entities are postulated; the learner simulator is a configuration of an existing LLM. The listed assumptions are the main unverified premises that the central claim depends on, and the listed free parameters are hand-set hyperparameters rather than constants fitted to outcomes.

free parameters (3)
  • fine-tuning epochs and batch size = 10 epochs, batch size 3
    Chosen for the Azure AI Studio fine-tune (§5.1); not fitted to test data, but affects the skill identifier's behavior.
  • LLM temperature = 0.7
    Set across all experiments (§5.1); a hand-chosen hyperparameter that could shift content quality and path diversity.
  • retrieved search results per query = 5
    Used by the RAG module (§5.1); hand-chosen to balance conciseness and coverage.
assumptions (4)
  • domain assumption Skills required for a job posting are a valid proxy for skills required to achieve a learner's goal.
    Goal-to-skill dataset and ground truth for evaluation are derived from job postings (§4.1.1, §5.2.2); if job summaries do not capture real learner goals, the mapping evaluation is off-target.
  • domain assumption An LLM learner simulator can produce feedback representative of real learners.
    The path scheduler and content creator rely on mimicked feedback from a role-playing LLM agent (§4.3.1) to refine resources without real user input.
  • domain assumption GPT-4o as an automated judge yields scores aligned with human judgment.
    Main automated evaluation uses GPT-4o as evaluator (§5.2); human validation covers only 2 experts, 20 samples, and Python-related results (§A.2).
  • domain assumption Self-reported engagement and satisfaction are acceptable proxies for learning effectiveness.
    The end-to-end human study measures subjective ratings, not learning gains or objective task completion (§6.3).

how reviews work

0 comments
Cite this review

Pith. "Pith review of LLM-powered Multi-agent Framework for Goal-oriented Learning in Intelligent Tutoring System." pith.science (2026). https://pith.science/paper/QZ6VEBNZ

@misc{pith2026250115749,
  author       = {Pith},
  title        = {Pith review of: LLM-powered Multi-agent Framework for Goal-oriented Learning in Intelligent Tutoring System},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QZ6VEBNZ}},
  note         = {Machine review of arXiv:2501.15749}
}
read the original abstract

Intelligent Tutoring Systems (ITSs) have revolutionized education by offering personalized learning experiences. However, as goal-oriented learning, which emphasizes efficiently achieving specific objectives, becomes increasingly important in professional contexts, existing ITSs often struggle to deliver this type of targeted learning experience. In this paper, we propose GenMentor, an LLM-powered multi-agent framework designed to deliver goal-oriented, personalized learning within ITS. GenMentor begins by accurately mapping learners' goals to required skills using a fine-tuned LLM trained on a custom goal-to-skill dataset. After identifying the skill gap, it schedules an efficient learning path using an evolving optimization approach, driven by a comprehensive and dynamic profile of learners' multifaceted status. Additionally, GenMentor tailors learning content with an exploration-drafting-integration mechanism to align with individual learner needs. Extensive automated and human evaluations demonstrate GenMentor's effectiveness in learning guidance and content quality. Furthermore, we have deployed it in practice and also implemented it as an application. Practical human study with professional learners further highlights its effectiveness in goal alignment and resource targeting, leading to enhanced personalization. Supplementary resources are available at https://github.com/GeminiLight/gen-mentor.

Figures

Figures reproduced from arXiv: 2501.15749 by the authors.

Figure 1
Figure 1. Comparison of three types of ITS Paradigms. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of the GenMentor: An LLM-powered multi-agent framework for goal-oriented learning in ITS. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. An illustration of dynamic learner modeling. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Human comparative preference. 6 END-TO-END HUMAN STUDY We have deployed GenMentor in practice and conducted the human study with professional learners for further evaluation. 6.1 Practical Deployment We have deployed GenMentor within our private product, the AIEP platf…
Figure 4
Figure 4. Figure 4: Evaluation results on created learning content. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: Questionnaire results from 20 participants (questions shortened for clarity). Gray texts are means and std. deviations. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: The core interfaces of GenMentor’s web application. Please see our supplementary resources for more details. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. Conversational Education at Scale: A Multi-LLM Agent Workflow for Procedural Learning and Pedagogic Quality Assessment

    cs.AI 2025-07 conditional novelty 5.0 of 10

    WikiHowAgent generates 114,296 simulated teacher-learner conversations from 14,287 WikiHow tutorials and evaluates their pedagogic quality with LLM and human judges.

  2. Partnering with AI: A Pedagogical Feedback System for LLM Integration into Programming Education

    cs.CY 2025-07 conditional novelty 5.0 of 10

    A multi-agent LLM feedback system for Python programming, built on pedagogical principles of mastery and progress, earned positive ratings from eight teachers though it cannot replace human context.

  3. Toward Personalizing Quantum Computing Education: An Evolutionary LLM-Powered Approach

    cs.CY 2025-04 conditional novelty 4.0 of 10

    A two-agent LLM tutoring system with a knowledge graph and a tag interface is proposed for personalized quantum computing education, demonstrated in a single simulated lesson.

Reference graph

Works this paper leans on

39 extracted references · 31 canonical work pages · cited by 3 Pith papers

  1. [1]

    Goal orientation and decision-making in education

    Vibe Aarkrog and Bjarne Wahlgren. Goal orientation and decision-making in education. Vocations and Learning, 15(1):71–86, 2022

    work page 2022

  2. [2]

    Learner modelling: systematic review of the literature from the last 5 years

    Abir Abyaa, Mohammed Khalidi Idrissi, and Samir Bennani. Learner modelling: systematic review of the literature from the last 5 years. Educational Technology Research and Development, 67:1105–1143, 2019

    work page 2019

  3. [3]

    Gpt-4 technical report

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Floren- cia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2023

    arXiv 2023

  4. [4]

    Intelligent Tutoring Systems: A Comprehensive Historical Survey with Recent Developments

    Ali Alkhatlan and Jugal Kalita. Intelligent tutoring systems: A comprehensive historical survey with recent developments. arXiv preprint arXiv:1812.09628, 2018

    work page Pith review arXiv 2018

  5. [5]

    Intent detection in the age of llms

    Gaurav Arora, Shreya Jain, and Srujana Merugu. Intent detection in the age of llms. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track , EMNLP, 2024

    work page 2024

  6. [6]

    Eportfolios for lifelong learning and assessment

    Darren Cambridge. Eportfolios for lifelong learning and assessment . John Wiley & Sons, 2010

    work page 2010

  7. [7]

    Empowering private tutoring by chaining large language models

    Yulin Chen, Ning Ding, Hai-Tao Zheng, Zhiyuan Liu, Maosong Sun, and Bowen Zhou. Empowering private tutoring by chaining large language models. In Pro- ceedings of the 32nd ACM International Conference on Information and Knowledge Management, CIKM, 2024

    work page 2024

  8. [8]

    Developing a Tutoring Dialog Dataset to Optimize LLMs for Educational Use

    Menna Fateen and Tsunenori Mine. Developing a tutoring dialog dataset to optimize llms for educational use. arXiv preprint arXiv:2410.19231, 2024

    work page Pith review arXiv 2024

Show all 39 references
  1. [9]

    Large language model based multi- agents: A survey of progress and challenges

    Taicheng Guo, Xiuying Chen, Yaqi Wang, Ruidi Chang, Shichao Pei, Nitesh V Chawla, Olaf Wiest, and Xiangliang Zhang. Large language model based multi- agents: A survey of progress and challenges. In Proceedings of the 33rd Interna- tional Joint Conference on Artificial Intellig...

    2024

  2. [10]

    Learnerexp: Exploring and explain- ing the time management of online learning activity

    Huan He, Qinghua Zheng, and Bo Dong. Learnerexp: Exploring and explain- ing the time management of online learning activity. In The World Wide Web Conference, WWW, page 3521–3525, 2019

    2019

  3. [11]

    Large language models struggle to learn long-tail knowledge

    Nikhil Kandpal, Haikang Deng, Adam Roberts, Eric Wallace, and Colin Raffel. Large language models struggle to learn long-tail knowledge. In International Conference on Machine Learning , ICML, pages 15696–15707, 2023

    2023

  4. [12]

    Prometheus 2: An open source language model specialized in evaluating other language models

    Seungone Kim, Juyoung Suk, Shayne Longpre, Bill Yuchen Lin, Jamin Shin, Sean Welleck, Graham Neubig, Moontae Lee, Kyungjae Lee, and Minjoon Seo. Prometheus 2: An open source language model specialized in evaluating other language models. In Proceedings of the 2024 Conference o...

    2024

  5. [13]

    Goal- oriented active learning (goal) system to promote reading engagement, self- directed learning behavior, and motivation in extensive reading

    Huiyong Li, Rwitajit Majumdar, Mei-Rong Alice Chen, and Hiroaki Ogata. Goal- oriented active learning (goal) system to promote reading engagement, self- directed learning behavior, and motivation in extensive reading. Computers & Education, 171:104239, 2021

    2021

  6. [14]

    Metaagents: Simulating interactions of human behaviors for llm-based task-oriented coordination via collaborative generative agents

    Yuan Li, Yixuan Zhang, and Lichao Sun. Metaagents: Simulating interactions of human behaviors for llm-based task-oriented coordination via collaborative generative agents. arXiv preprint arXiv:2310.06500, 2023

    2023 arXiv

  7. [15]

    Exploiting cognitive structure for adaptive learning

    Qi Liu, Shiwei Tong, Chuanren Liu, Hongke Zhao, Enhong Chen, Haiping Ma, and Shijin Wang. Exploiting cognitive structure for adaptive learning. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD, page 627–635, 2019

    2019

  8. [16]

    Personality-aware student simulation for conversational intelligent tutoring systems

    Zhengyuan Liu, Stella Xin Yin, Geyu Lin, and Nancy F Chen. Personality-aware student simulation for conversational intelligent tutoring systems. arXiv preprint arXiv:2404.06762, 2024

    2024 arXiv

  9. [17]

    Opportunities and challenges in neural dialog tutoring

    Jakub Macina, Nico Daheim, Lingzhi Wang, Tanmay Sinha, Manu Kapur, Iryna Gurevych, and Mrinmaya Sachan. Opportunities and challenges in neural dialog tutoring. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics , EACL...

    2023

  10. [18]

    Roco: Dialectic multi-robot collab- oration with large language models

    Zhao Mandi, Shreeya Jain, and Shuran Song. Roco: Dialectic multi-robot collab- oration with large language models. In 2024 IEEE International Conference on Robotics and Automation, ICRA, pages 286–299, 2024

    2024

  11. [19]

    Intelligent tutoring systems: an overview

    Hyacinth S Nwana. Intelligent tutoring systems: an overview. Artificial Intelli- gence Review, 4(4):251–277, 1990

    1990

  12. [20]

    Generative large language models for dialog-based tutoring: An early consideration of opportunities and concerns

    Benjamin D Nye, Dillon Mee, and Mark G Core. Generative large language models for dialog-based tutoring: An early consideration of opportunities and concerns. In LLM@ AIED, pages 78–88, 2023

    2023

  13. [21]

    Generative agents: Interactive simulacra of human behavior

    Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th annual acm symposium on user interface software and technology , UIST, pages 1–22, 2023

    2023

  14. [22]

    Empowering personalized learning through a conversation-based tutoring system with student modeling

    Minju Park, Sojung Kim, Seunghyun Lee, Soonwoo Kwon, and Kyuseok Kim. Empowering personalized learning through a conversation-based tutoring system with student modeling. In Extended Abstracts of the CHI Conference on Human Factors in Computing Systems , CHI, pages 1–10, 2024

    2024

  15. [23]

    Chatdev: Communicative agents for software development

    Chen Qian, Wei Liu, Hongzhang Liu, Nuo Chen, Yufan Dang, Jiahao Li, Cheng Yang, Weize Chen, Yusheng Su, Xin Cong, et al. Chatdev: Communicative agents for software development. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1...

    2024

  16. [24]

    Role play with large language models

    Murray Shanahan, Kyle McDonell, and Laria Reynolds. Role play with large language models. Nature, 623(7987):493–498, 2023

    2023

  17. [25]

    Assisting in writing wikipedia-like articles from scratch with large language models

    Yijia Shao, Yucheng Jiang, Theodore A Kanell, Peter Xu, Omar Khattab, and Monica S Lam. Assisting in writing wikipedia-like articles from scratch with large language models. In Proceedings of the 2024 Conference of the North Ameri- can Chapter of the Association for Computatio...

    2024

  18. [26]

    De- sign recommendations for intelligent tutoring systems: Volume 1-learner modeling , volume 1

    Robert A Sottilare, Arthur Graesser, Xiangen Hu, and Heather Holden. De- sign recommendations for intelligent tutoring systems: Volume 1-learner modeling , volume 1. US Army Research Laboratory, 2013

    2013

  19. [27]

    Interpretable knowledge tracing with multiscale state representation

    Jianwen Sun, Fenghua Yu, Qian Wan, Qing Li, Sannyuya Liu, and Xiaoxuan Shen. Interpretable knowledge tracing with multiscale state representation. In Proceedings of the ACM Web Conference 2024 , WWW, page 3265–3276, 2024

    2024

  20. [28]

    Measuring cognitive load

    John Sweller, Paul Ayres, Slava Kalyuga, John Sweller, Paul Ayres, and Slava Kalyuga. Measuring cognitive load. Cognitive load theory, pages 71–85, 2011

    2011

  21. [29]

    Learner modeling in adaptive educational systems: A comparative study

    Mouenis Anouar Tadlaoui, Souhaib Aammou, Mohamed Khaldi, and Rommel No- vaes Carvalho. Learner modeling in adaptive educational systems: A comparative study. International Journal of Modern Education & Computer Science , 8(3), 2016

    2016

  22. [30]

    Incremental cognitive diagnosis for intelligent education

    Shiwei Tong, Jiayu Liu, Yuting Hong, Zhenya Huang, Le Wu, Qi Liu, Wei Huang, Enhong Chen, and Dan Zhang. Incremental cognitive diagnosis for intelligent education. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , KDD, pages 1760–1770, 2022

    2022

  23. [31]

    Llama: Open and efficient foundation language models

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023

    2023 arXiv

  24. [32]

    Ten steps to complex learning: A systematic approach to four-component instructional design

    Jeroen JG Van Merriënboer and Paul A Kirschner. Ten steps to complex learning: A systematic approach to four-component instructional design . Routledge, 2017

    2017

  25. [33]

    Large language models for education: A survey and outlook

    Shen Wang, Tianlong Xu, Hang Li, Chaoli Zhang, Joleen Liang, Jiliang Tang, Philip S Yu, and Qingsong Wen. Large language models for education: A survey and outlook. arXiv preprint arXiv:2403.18105, 2024

    2024 arXiv

  26. [34]

    Chain-of-thought prompting elicits reasoning in large language models

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems , 35:24824–24837, 2022

    2022

  27. [35]

    A critical evaluation of evaluations for long-form question answering

    Fangyuan Xu, Yixiao Song, Mohit Iyyer, and Eunsol Choi. A critical evaluation of evaluations for long-form question answering. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , ACL, 2023

    2023

  28. [36]

    Language agents with reinforcement learning for strategic play in the werewolf game

    Zelai Xu, Chao Yu, Fei Fang, Yu Wang, and Yi Wu. Language agents with reinforcement learning for strategic play in the werewolf game. arXiv preprint arXiv:2310.18940, 2023

    2023 arXiv

  29. [37]

    Doubly constrained offline reinforcement learning for learning path recommendation

    Yue Yun, Huan Dai, Rui An, Yupei Zhang, and Xuequn Shang. Doubly constrained offline reinforcement learning for learning path recommendation. Knowledge- Based Systems, page 111242, 2023

    2023

  30. [38]

    Judging llm-as-a-judge with mt-bench and chatbot arena

    Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, et al. Judging llm-as-a-judge with mt-bench and chatbot arena. Advances in Neural Information Processing Systems, 36, 2023

    2023

  31. [39]

    A robust computerized adaptive testing approach in educational question retrieval

    Yan Zhuang, Qi Liu, Zhenya Huang, Zhi Li, Binbin Jin, Haoyang Bi, Enhong Chen, and Shijin Wang. A robust computerized adaptive testing approach in educational question retrieval. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Infor...

    2022

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.