pith. sign in

arxiv: 2606.20287 · v1 · pith:5GYPWPUHnew · submitted 2026-06-18 · 💻 cs.CL

PsyScore: A Psychometrically-Aware Framework for Trait-Adaptive Essay Scoring and ZPD-Scaffolded Feedback

Wei Xia , Jin Wu , Haoran Shi , Xiangyu Wang , Chanjin Zheng This is my paper

Pith reviewed 2026-06-26 17:36 UTC · model grok-4.3

classification 💻 cs.CL
keywords automated essay scoringitem response theoryadaptive feedbackpsychometric modelingzone of proximal developmentlarge language modelseducational assessment
0
0 comments X

The pith

PsyScore links essay scoring to ability-adapted feedback through a shared psychometric parameter.

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

The paper introduces PsyScore to overcome the split between accurate automated essay scoring and feedback that actually matches a learner's current level. It does this by deriving one latent ability value from a neural implementation of the graded partial credit model and then using that same value to steer a multi-agent feedback generator toward zone-of-proximal-development scaffolds. On the ASAP++ dataset the resulting scores stay competitive with existing models while the generated feedback receives higher ratings for pedagogical fit in both pairwise preference tests and simulated revision tasks.

Core claim

PsyScore comprises a Trait-Adaptive Neural IRT Scorer that embeds the Graded Partial Credit Model to produce both essay scores and an interpretable ability parameter, a ZPD-Scaffolded Feedback Generator that conditions multi-agent strategies on that parameter, and a Multi-Perspective Feedback Evaluation Strategy that measures quality through preference judgments and revision simulations; the shared ability representation thereby unifies diagnostic assessment with level-specific instructional support.

What carries the argument

The shared latent ability parameter produced by the Trait-Adaptive Neural IRT Scorer, which directly conditions the ZPD-Scaffolded Feedback Generator.

If this is right

  • Scoring accuracy on ASAP++ remains competitive with prior neural AES systems.
  • Feedback becomes more aligned with learner proficiency as judged by preference and revision metrics.
  • A single ability estimate supports both assessment and scaffolding without separate models.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same ability-conditioned loop could be tested on short-answer or programming tasks to check whether the unification generalizes.
  • If the ability parameter proves stable across multiple essay prompts, the framework offers a route to longitudinal tracking of skill growth inside one system.

Load-bearing premise

The ability parameter estimated by the Trait-Adaptive Neural IRT Scorer can be directly used to condition multi-agent feedback strategies so that instructional focus adapts effectively across proficiency levels.

What would settle it

A controlled trial in which feedback generated without conditioning on the estimated ability parameter yields equal or higher student revision quality and preference scores than the ability-conditioned version.

Figures

Figures reproduced from arXiv: 2606.20287 by Chanjin Zheng, Haoran Shi, Jin Wu, Wei Xia, Xiangyu Wang.

Figure 1
Figure 1. Figure 1: Comparison of traditional approaches and [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the PsyScore framework. (a) Trait-Adaptive GPCM Scorer estimates the student’s latent ability (θ) and outputs a diagnostic vector (Dx). (b) ZPD-Conditional Feedback Generator synthesizes consensus feedback (ff inal) by mapping θ to adaptive strategies via multi-agent fusion. (c) Multi-Perspective Evaluation validates quality via intrinsic LLM-based comparison and extrinsic simulated revision. a… view at source ↗
Figure 3
Figure 3. Figure 3: Pairwise preference evaluation results across four baselines. The bars represent the number of wins awarded by judges. PsyScore demonstrates consistent superiority across both open-source (a-c) and closed-source (d) models, particularly in Actionability and Adaptivity [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
read the original abstract

Effective Automated Essay Scoring (AES) are expected to support both reliable assessment and actionable instructional feedback. However, existing approaches often treat scoring and feedback as separate components: neural scoring models provide limited interpretability, while Large Language Model (LLM)-based feedback is typically insensitive to learners proficiency levels. To address this fragmentation, this work proposes PsyScore, a psychometrically-aware framework that integrates diagnostic assessment with instructional scaffolding through a shared latent ability representation. PsyScore comprises three key modules: a Trait-Adaptive Neural IRT Scorer that incorporates the Graded Partial Credit Model (GPCM) into a neural architecture, enabling the precise estimation of student ability while maintaining psychometric interpretability, a ZPD-Scaffolded Feedback Generator, which conditions multi-agent feedback strategies on the diagnosed ability parameter to adapt instructional focus across different proficiency levels, and a Multi-Perspective Feedback Evaluation Strategy that assesses feedback quality via pairwise preference judgements and student revision simulations. Experiments on the ASAP++ dataset demonstrate that PsyScore achieves competitive scoring performance while providing more pedagogically aligned feedback.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript proposes PsyScore, a framework integrating three modules: a Trait-Adaptive Neural IRT Scorer that embeds the Graded Partial Credit Model (GPCM) into a neural architecture for essay scoring and student ability estimation, a ZPD-Scaffolded Feedback Generator that conditions multi-agent LLM feedback strategies on the estimated ability parameter to adapt instructional focus, and a Multi-Perspective Feedback Evaluation Strategy that uses pairwise preference judgments and student revision simulations to assess feedback quality. Experiments on the ASAP++ dataset are reported to show competitive scoring performance alongside more pedagogically aligned feedback than prior approaches.

Significance. If the central results hold after addressing the evaluation gap, the work would offer a concrete bridge between psychometric models and LLM-based feedback systems, potentially improving interpretability and adaptivity in automated essay scoring. The shared latent ability representation is a clear conceptual strength that could influence future designs in educational NLP if the conditioning effect is isolated and quantified.

major comments (1)
  1. [Multi-Perspective Feedback Evaluation Strategy] Multi-Perspective Feedback Evaluation Strategy: the reported experiments do not include an ablation isolating the effect of conditioning the ZPD-Scaffolded Feedback Generator on the GPCM-derived ability parameter (e.g., ability-conditioned multi-agent feedback versus fixed-prompt or unconditioned baselines). This comparison is required to substantiate the claim that the shared latent representation yields measurably more pedagogically aligned output or revision gains; without it, improvements cannot be attributed to the ability-parameter mechanism rather than other design choices.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive comment on evaluation design. We agree that isolating the contribution of the ability-parameter conditioning is necessary to strengthen the central claim and will add the requested ablation in revision.

read point-by-point responses

  1. Referee: [Multi-Perspective Feedback Evaluation Strategy] Multi-Perspective Feedback Evaluation Strategy: the reported experiments do not include an ablation isolating the effect of conditioning the ZPD-Scaffolded Feedback Generator on the GPCM-derived ability parameter (e.g., ability-conditioned multi-agent feedback versus fixed-prompt or unconditioned baselines). This comparison is required to substantiate the claim that the shared latent representation yields measurably more pedagogically aligned output or revision gains; without it, improvements cannot be attributed to the ability-parameter mechanism rather than other design choices.

    Authors: We agree that the current experiments do not contain a direct ablation isolating the conditioning of the ZPD-Scaffolded Feedback Generator on the GPCM-derived ability parameter. The manuscript reports overall competitive scoring and feedback alignment but does not compare ability-conditioned multi-agent strategies against fixed-prompt or unconditioned baselines. In the revised version we will add this ablation on the ASAP++ dataset, reporting pairwise preference judgments and revision-simulation outcomes for the three conditions. This will allow quantification of the incremental effect attributable to the shared latent ability representation. revision: yes

Circularity Check

0 steps flagged

No circularity: framework integrates independent modules via shared latent variable

full rationale

The derivation chain estimates student ability via the Trait-Adaptive Neural IRT Scorer (incorporating GPCM), then uses that parameter to condition the separate ZPD-Scaffolded Feedback Generator, and evaluates the output with an independent Multi-Perspective Feedback Evaluation Strategy (pairwise preferences and revision simulations). None of these steps reduce to self-definition, fitted-input-as-prediction, or self-citation load-bearing; the shared latent representation is an explicit design choice rather than a tautology, and no equations or claims in the abstract or described modules collapse the output back to the input by construction. The paper remains self-contained against external benchmarks on ASAP++.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The framework rests on standard IRT modeling assumptions and LLM generation capabilities. The central ability parameter is a fitted latent variable whose quality determines both scoring and feedback.

free parameters (1)
  • student ability parameter
    Latent trait estimated from essay responses via the neural GPCM model; used for both scoring and feedback conditioning.
axioms (1)
  • domain assumption The Graded Partial Credit Model can be incorporated into a neural architecture while preserving psychometric interpretability.
    Invoked as the basis for the Trait-Adaptive Neural IRT Scorer.

pith-pipeline@v0.9.1-grok · 5725 in / 1090 out tokens · 45093 ms · 2026-06-26T17:36:26.707932+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

170 extracted references · 72 canonical work pages

  1. [1]

    Aho and Jeffrey D

    Alfred V. Aho and Jeffrey D. Ullman , title =. 1972

    1972

  2. [2]

    Publications Manual , year = "1983", publisher =

    1983

  3. [3]

    Chandra and Dexter C

    Ashok K. Chandra and Dexter C. Kozen and Larry J. Stockmeyer , year = "1981", title =. doi:10.1145/322234.322243

    work page doi:10.1145/322234.322243 1981

  4. [4]

    Scalable training of

    Andrew, Galen and Gao, Jianfeng , booktitle=. Scalable training of

  5. [5]

    Dan Gusfield , title =. 1997

    1997

  6. [6]

    Tetreault , title =

    Mohammad Sadegh Rasooli and Joel R. Tetreault , title =. Computing Research Repository , volume =. 2015 , url =

    2015

  7. [7]

    A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data , Volume =

    Ando, Rie Kubota and Zhang, Tong , Issn =. A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data , Volume =. Journal of Machine Learning Research , Month = dec, Numpages =

  8. [8]

    Dual-scale

    Cho, Minsoo and Huang, Jin-Xia and Kwon, Oh-Woog , year =. Dual-scale. ETRI Journal , volume =

  9. [10]

    2025 , issn =

    An interpretable polytomous cognitive diagnosis framework for predicting examinee performance , journal =. 2025 , issn =. doi:https://doi.org/10.1016/j.ipm.2024.103913 , url =

    work page doi:10.1016/j.ipm.2024.103913 2025

  10. [11]

    2025 , eprint=

    MetaCD: A Meta Learning Framework for Cognitive Diagnosis based on Continual Learning , author=. 2025 , eprint=

    2025

  11. [12]

    2020 , publisher =

    A Trait-based Deep Learning Automated Essay Scoring System with Adaptive Feedback , journal =. 2020 , publisher =. doi:10.14569/IJACSA.2020.0110538 , url =

    work page doi:10.14569/ijacsa.2020.0110538 2020

  12. [13]

    Unleashing

    Lee, Sanwoo and Cai, Yida and Meng, Desong and Wang, Ziyang and Wu, Yunfang , year =. Unleashing. doi:10.48550/arXiv.2404.04941 , note =

    work page doi:10.48550/arxiv.2404.04941

  13. [15]

    doi:10.48550/ARXIV.2502.11916 , note =

    Su, Jiamin and Yan, Yibo and Fu, Fangteng and Zhang, Han and Ye, Jingheng and Liu, Xiang and Huo, Jiahao and Zhou, Huiyu and Hu, Xuming , year =. doi:10.48550/ARXIV.2502.11916 , note =

    work page doi:10.48550/arxiv.2502.11916

  14. [16]

    T - MES : Trait-Aware Mix-of-Experts Representation Learning for Multi-trait Essay Scoring

    Wang, Jiong and Liu, Jie. T - MES : Trait-Aware Mix-of-Experts Representation Learning for Multi-trait Essay Scoring. Proceedings of the 31st International Conference on Computational Linguistics. 2025

    2025

  15. [17]

    Automated

    Ormerod, Christopher , month =. Automated. 2025 , note =. doi:10.48550/arXiv.2505.22771 , publisher =

    work page doi:10.48550/arxiv.2505.22771 2025

  16. [18]

    2025 , note =

    Jordan, Joaquin and Yin, Xavier and Fabros, Melissa and Ranade, Gireeja and Norouzi, Narges , month =. 2025 , note =. doi:10.48550/arXiv.2506.13037 , publisher =

    work page doi:10.48550/arxiv.2506.13037 2025

  17. [19]

    Zafar, Samra and Minhas, Shaheer and Zaidi, Syed Ali Hassan and Naeem, Arfa and Ali, Zahra , month =. ". 2025 , note =. doi:10.48550/arXiv.2506.08221 , publisher =

    work page doi:10.48550/arxiv.2506.08221 2025

  18. [20]

    doi:10.1145/3726302.3730143 , author =

    2025 , note =. doi:10.1145/3726302.3730143 , author =

    work page doi:10.1145/3726302.3730143 2025

  19. [22]

    Transfer

    Morris, Oscar , month =. Transfer. 2025 , note =. doi:10.48550/arXiv.2503.11836 , publisher =

    work page doi:10.48550/arxiv.2503.11836 2025

  20. [23]

    2025 , note =

    Liu, Zhexiong and Litman, Diane and Wang, Elaine and Li, Tianwen and Gobat, Mason and Matsumura, Lindsay Clare and Correnti, Richard , month =. 2025 , note =. doi:10.48550/arXiv.2501.00715 , publisher =

    work page doi:10.48550/arxiv.2501.00715 2025

  21. [24]

    Advancing

    Zeinalipour, Kamyar and Mehak, Mehak and Parsamotamed, Fatemeh and Maggini, Marco and Gori, Marco , month =. Advancing. 2025 , note =. doi:10.48550/arXiv.2501.07740 , publisher =

    work page doi:10.48550/arxiv.2501.07740 2025

  22. [25]

    2024 , note =

    Kim, Minsun and Kim, SeonGyeom and Lee, Suyoun and Yoon, Yoosang and Myung, Junho and Yoo, Haneul and Lim, Hyunseung and Han, Jieun and Kim, Yoonsu and Ahn, So-Yeon and Kim, Juho and Oh, Alice and Hong, Hwajung and Lee, Tak Yeon , month =. 2024 , note =. doi:10.48550/arXiv.2410.15025 , publisher =

    work page doi:10.48550/arxiv.2410.15025 2024

  23. [26]

    Wang, Yupei and Hu, Renfen and Zhao, Zhe , month =. Beyond. 2024 , note =. doi:10.48550/arXiv.2405.19433 , publisher =

    work page doi:10.48550/arxiv.2405.19433 2024

  24. [27]

    Investigating

    Katuka, Gloria Ashiya and Gain, Alexander and Yu, Yen-Yun , month =. Investigating. 2024 , note =. doi:10.48550/arXiv.2405.00602 , publisher =

    work page doi:10.48550/arxiv.2405.00602 2024

  25. [28]

    , month =

    Karizaki, Mahsa Sheikhi and Gnesdilow, Dana and Puntambekar, Sadhana and Passonneau, Rebecca J. , month =. How. 2024 , note =. doi:10.48550/arXiv.2404.11682 , publisher =

    work page doi:10.48550/arxiv.2404.11682 2024

  26. [29]

    Wang, Izia Xiaoxiao and Wu, Xihan and Coates, Edith and Zeng, Min and Kuang, Jiexin and Liu, Siliang and Qiu, Mengyang and Park, Jungyeul , month =. Neural. 2024 , note =. doi:10.48550/arXiv.2402.17613 , publisher =

    work page doi:10.48550/arxiv.2402.17613 2024

  27. [30]

    , month =

    Yoon, Su-Youn and Miszoglad, Eva and Pierce, Lisa R. , month =. Evaluation of. 2023 , note =. doi:10.48550/arXiv.2310.06505 , publisher =

    work page doi:10.48550/arxiv.2310.06505 2023

  28. [31]

    2023 , note =

    Solopova, Veronika and Gruszczynski, Adrian and Rostom, Eiad and Cremer, Fritz and Witte, Sascha and Zhang, Chengming and Plößl, Fernando Ramos López Lea and Hofmann, Florian and Romeike, Ralf and Gläser-Zikuda, Michaela and Benzmüller, Christoph and Landgraf, Tim , month =. 2023 , note =. doi:10.48550/arXiv.2307.07523 , publisher =

    work page doi:10.48550/arxiv.2307.07523 2023

  29. [32]

    Review of feedback in

    Jong, You-Jin and Kim, Yong-Jin and Ri, Ok-Chol , month =. Review of feedback in. 2023 , note =. doi:10.48550/arXiv.2307.05553 , publisher =

    work page doi:10.48550/arxiv.2307.05553 2023

  30. [33]

    Predicting the

    Liu, Zhexiong and Litman, Diane and Wang, Elaine and Matsumura, Lindsay and Correnti, Richard , month =. Predicting the. 2023 , note =. doi:10.48550/arXiv.2306.00667 , publisher =

    work page doi:10.48550/arxiv.2306.00667 2023

  31. [34]

    British Journal of Educational Technology , author =

    Practical and. British Journal of Educational Technology , author =. 2024 , note =. doi:10.1111/bjet.13370 , number =

    work page doi:10.1111/bjet.13370 2024

  32. [35]

    Transformer

    Abhishek, Tushar and Rawat, Daksh and Gupta, Manish and Varma, Vasudeva , month =. Transformer. 2022 , note =. doi:10.48550/arXiv.2109.02176 , publisher =

    work page doi:10.48550/arxiv.2109.02176 2022

  33. [36]

    Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems , articleno =

    Afrin, Tazin and Kashefi, Omid and Olshefski, Christopher and Litman, Diane and Hwa, Rebecca and Godley, Amanda , month =. Effective. 2021 , note =. doi:10.1145/3411764.3445683 , booktitle =

    work page doi:10.1145/3411764.3445683 2021

  34. [37]

    Hong, Shengxin and Cai, Chang and Du, Sixuan and Feng, Haiyue and Liu, Siyuan and Fan, Xiuyi , month =. ". 2024 , note =. doi:10.48550/arXiv.2409.07453 , publisher =

    work page doi:10.48550/arxiv.2409.07453 2024

  35. [38]

    2025 , note =

    Shibata, Takumi and Miyamura, Yuichi , month =. 2025 , note =. doi:10.48550/arXiv.2505.08498 , publisher =

    work page doi:10.48550/arxiv.2505.08498 2025

  36. [39]

    Operationalizing

    Plasencia-Calaña, Yenisel , month =. Operationalizing. 2025 , note =. doi:10.48550/arXiv.2506.21603 , publisher =

    work page doi:10.48550/arxiv.2506.21603 2025

  37. [40]

    Yoshida, Lui , month =. Do. 2025 , note =. doi:10.48550/arXiv.2505.01035 , publisher =

    work page doi:10.48550/arxiv.2505.01035 2025

  38. [41]

    and Kwong, Theresa and Atif, Amara , month =

    Kamalov, Firuz and Calonge, David Santandreu and Smail, Linda and Azizov, Dilshod and Thadani, Dimple R. and Kwong, Theresa and Atif, Amara , month =. Evolution of. 2025 , note =. doi:10.48550/arXiv.2504.20082 , publisher =

    work page doi:10.48550/arxiv.2504.20082 2025

  39. [42]

    Cai, Yida and Liang, Kun and Lee, Sanwoo and Wang, Qinghan and Wu, Yunfang , month =. Rank-. 2025 , note =. doi:10.48550/arXiv.2504.05736 , publisher =

    work page doi:10.48550/arxiv.2504.05736 2025

  40. [43]

    and Yang, Yi and Abbasi, Ahmed , month =

    Oketch, Kezia and Lalor, John P. and Yang, Yi and Abbasi, Ahmed , month =. Bridging the. 2025 , note =. doi:10.48550/arXiv.2503.11827 , publisher =

    work page doi:10.48550/arxiv.2503.11827 2025

  41. [44]

    Teach-to-

    Do, Heejin and Ryu, Sangwon and Lee, Gary Geunbae , month =. Teach-to-. 2025 , note =. doi:10.48550/arXiv.2502.20748 , publisher =

    work page doi:10.48550/arxiv.2502.20748 2025

  42. [45]

    How well can

    Ghazawi, Rayed and Simpson, Edwin , month =. How well can. 2025 , note =. doi:10.48550/arXiv.2501.16516 , publisher =

    work page doi:10.48550/arxiv.2501.16516 2025

  43. [46]

    Wendlinger, Lorenz and Braun, Christian and Zubaer, Abdullah Al and Nonn, Simon Alexander and Großkopf, Sarah and Fellicious, Christofer and Granitzer, Michael , month =. On the. 2024 , note =. doi:10.48550/arXiv.2412.15902 , publisher =

    work page doi:10.48550/arxiv.2412.15902 2024

  44. [47]

    Evaluating

    Zhong, Yang and Hao, Jiangang and Fauss, Michael and Li, Chen and Wang, Yuan , month =. Evaluating. 2024 , note =. doi:10.48550/arXiv.2410.17439 , publisher =

    work page doi:10.48550/arxiv.2410.17439 2024

  45. [48]

    Kundu, Anindita and Barbosa, Denilson , month =. Are. 2024 , note =. doi:10.48550/arXiv.2409.13120 , publisher =

    work page doi:10.48550/arxiv.2409.13120 2024

  46. [49]

    Kim, Seungju and Jo, Meounggun , month =. Is. 2024 , note =. doi:10.1145/3657604.3664703 , booktitle =

    work page doi:10.1145/3657604.3664703 2024

  47. [50]

    Attention-based Recurrent Convolutional Neural Network for Automatic Essay Scoring

    Dong, Fei and Zhang, Yue and Yang, Jie , year =. Attention-based. doi:10.18653/v1/K17-1017 , booktitle =

    work page doi:10.18653/v1/k17-1017

  48. [51]

    Many Hands Make Light Work: Using Essay Traits to Automatically Score Essays , booktitle =

    Rahul Kumar and Sandeep Mathias and Sriparna Saha and Pushpak Bhattacharyya , editor =. Many Hands Make Light Work: Using Essay Traits to Automatically Score Essays , booktitle =. 2022 , url =

    2022

  49. [52]

    Automated essay scoring with string kernels and word embeddings

    Cozma, M a d a lina and Butnaru, Andrei and Ionescu, Radu Tudor. Automated essay scoring with string kernels and word embeddings. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 2018. doi:10.18653/v1/P18-2080

    work page doi:10.18653/v1/p18-2080 2018

  50. [53]

    2023 , doi =

    Document. 2023 , doi =

    2023

  51. [54]

    Linking essay-writing tests using many-facet models and neural automated essay scoring , url =

    Uto, Masaki and Aramaki, Kota , month =. Linking essay-writing tests using many-facet models and neural automated essay scoring , url =. doi:10.3758/s13428-024-02485-2 , journal =

    work page doi:10.3758/s13428-024-02485-2

  52. [55]

    Uto, Masaki and Takahashi, Yuto , editor =. Neural. Artificial. 2024 , doi =

    2024

  53. [56]

    Difficulty-

    Tomikawa, Yuto and Uto, Masaki , editor =. Difficulty-. Artificial. 2024 , doi =

    2024

  54. [57]

    Artificial

    Shindo, Naoki and Uto, Masaki , editor =. Artificial. 2024 , doi =

    2024

  55. [58]

    Behavior Research Methods , author =

    A. Behavior Research Methods , author =. 2022 , pages =. doi:10.3758/s13428-022-01997-z , number =

    work page doi:10.3758/s13428-022-01997-z 2022

  56. [59]

    Yamaura, Misato and Fukuda, Itsuki and Uto, Masaki , editor =. Neural. Artificial. 2023 , doi =

    2023

  57. [60]

    Behavior Research Methods , author =

    Accuracy of performance-test linking based on a many-facet. Behavior Research Methods , author =. 2021 , pages =. doi:10.3758/s13428-020-01498-x , number =

    work page doi:10.3758/s13428-020-01498-x 2021

  58. [61]

    Behaviormetrika , author =

    Special issue: e-testing from artificial intelligence approach , volume =. Behaviormetrika , author =. 2021 , pages =. doi:10.1007/s41237-021-00146-8 , number =

    work page doi:10.1007/s41237-021-00146-8 2021

  59. [62]

    Behaviormetrika , author =

    A multidimensional generalized many-facet. Behaviormetrika , author =. 2021 , pages =. doi:10.1007/s41237-021-00144-w , number =

    work page doi:10.1007/s41237-021-00144-w 2021

  60. [63]

    Behaviormetrika , author =

    A review of deep-neural automated essay scoring models , volume =. Behaviormetrika , author =. 2021 , pages =. doi:10.1007/s41237-021-00142-y , number =

    work page doi:10.1007/s41237-021-00142-y 2021

  61. [64]

    Integration of

    Aomi, Itsuki and Tsutsumi, Emiko and Uto, Masaki and Ueno, Maomi , editor =. Integration of. Artificial. 2021 , doi =

    2021

  62. [65]

    Uto, Masaki , editor =. A. Artificial. 2021 , doi =

    2021

  63. [66]

    Estimating

    Nakayama, Minoru and Sciarrone, Filippo and Uto, Masaki and Temperini, Marco , editor =. Estimating. Methodologies and. 2021 , doi =

    2021

  64. [67]

    Behaviormetrika , author =

    A generalized many-facet. Behaviormetrika , author =. 2020 , pages =. doi:10.1007/s41237-020-00115-7 , number =

    work page doi:10.1007/s41237-020-00115-7 2020

  65. [68]

    International Journal of Artificial Intelligence in Education , author =

    Time- and. International Journal of Artificial Intelligence in Education , author =. 2020 , pages =. doi:10.1007/s40593-019-00189-9 , number =

    work page doi:10.1007/s40593-019-00189-9 2020

  66. [69]

    Automated

    Uto, Masaki and Uchida, Yuto , editor =. Automated. Artificial. 2020 , doi =

    2020

  67. [70]

    Uto, Masaki and Okano, Masashi , editor =. Robust. Artificial. 2020 , doi =

    2020

  68. [71]

    Uto, Masaki , editor =. Rater-. Artificial. 2019 , doi =

    2019

  69. [72]

    Social constructivist approach of motivation: social media messages recommendation system , url =

    Louvigné, Sébastien and Uto, Masaki and Kato, Yoshihiro and Ishii, Takatoshi , month =. Social constructivist approach of motivation: social media messages recommendation system , url =. doi:10.1007/s41237-017-0043-7 , journal =

    work page doi:10.1007/s41237-017-0043-7

  70. [73]

    Uto, Masaki and Ueno, Maomi , editor =. Item. Artificial. 2018 , doi =

    2018

  71. [74]

    and Yancey, Kevin and von Davier, Alina A

    Burstein, Jill and LaFlair, Geoffrey T. and Yancey, Kevin and von Davier, Alina A. and Dotan, Ravit , year =. Responsible. doi:10.48550/ARXIV.2409.07476 , publisher =

    work page doi:10.48550/arxiv.2409.07476

  72. [75]

    Collaborative

    Aramaki, Kota and Uto, Masaki , editor =. Collaborative. Artificial. 2024 , doi =

    2024

  73. [76]

    Ridley, Robert and He, Liang and Dai, Xinyu and Huang, Shujian and Chen, Jiajun , month =. Prompt. 2020 , note =

    2020

  74. [77]

    Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2024) , month =

    Automated Essay Scoring Using Grammatical Variety and Errors with Multi-Task Learning and Item Response Theory , author =. Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2024) , month =. 2024 , address =

    2024

  75. [78]

    IEEE Transactions on Learning Technologies , author =

    Integration of. IEEE Transactions on Learning Technologies , author =. 2023 , pages =. doi:10.1109/TLT.2023.3253215 , number =

    work page doi:10.1109/tlt.2023.3253215 2023

  76. [79]

    IEEE Transactions on Learning Technologies , author =

    Learning. IEEE Transactions on Learning Technologies , author =. 2021 , pages =. doi:10.1109/TLT.2022.3145352 , number =

    work page doi:10.1109/tlt.2022.3145352 2021

  77. [80]

    Analytic

    Shibata, Takumi and Uto, Masaki , year =. Analytic

  78. [81]

    doi:https://doi.org/10.1016/j.eswa.2023.123043 , journal =

    Liu, Yuanchao and Han, Jiawei and Sboev, Alexander and Makarov, Ilya , year =. doi:https://doi.org/10.1016/j.eswa.2023.123043 , journal =

    work page doi:10.1016/j.eswa.2023.123043 2023

  79. [82]

    1990 , publisher =

    Item Response Theory , author =. 1990 , publisher =

    1990

  80. [83]

    1991 , publisher =

    Fundamentals of Item Response Theory , author =. 1991 , publisher =

    1991

Showing first 80 references.