{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T06:09:14Z","timestamp":1789711754953,"version":"4.0.1"},"reference-count":44,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T00:00:00Z","timestamp":1789689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Automated Essay Scoring (AES) has emerged as an important application of natural language processing providing scalable and consistent evaluation of written responses. These assessments can be conducted using holistic or analytic evaluation approaches. Recent studies have heavily focused on utilizing deep learning models and particularly transformer-based models in holistic essay scoring, while accurately modeling the multi-dimensional analytical scoring remains a challenging problem. Despite their strong performance in holistic essay scoring, most existing approaches work as black-box predictors and do not explicitly model the reasoning process underlying analytic rubric-based evaluation. In this paper, we propose ARGUE-TIRA, a multi-task framework for joint holistic and analytic essay scoring. The proposed framework first predicts multiple analytic writing traits and subsequently applies a Trait Interaction Reasoning Attention (TIRA) to explicitly model dependencies among rubric dimensions before generating the final holistic score. Unlike conventional multi-task AES approaches that independently predict analytic traits, the proposed reasoning mechanism captures inter-trait relationships through structured self-attention reasoning, improving both scoring consistency and interpretability. Experiments conducted on the ASAP and ASAP++ datasets demonstrate that ARGUE-TIRA achieves competitive holistic performance and strong analytic trait-scoring performance. The framework obtains an average holistic QWK of 0.762 on ASAP and the highest average analytic QWK of 0.7424 on ASAP++. Furthermore, it achieves the best performance in seven of the eight evaluated analytic traits, including content, organization, word choice, sentence fluency, conventions, prompt adherence, and language quality.<\/jats:p>","DOI":"10.3389\/frai.2026.1895359","type":"journal-article","created":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T05:36:05Z","timestamp":1789709765000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["ARGUE-TIRA: attention-based trait interaction reasoning for holistic and analytic automated essay scoring"],"prefix":"10.3389","volume":"9","author":[{"given":"Areej","family":"Alhothali","sequence":"first","affiliation":[{"name":"Department of Computer Science, Faculty of Computing and Information Technology, King Abdulaziz University","place":["Jeddah, Saudi Arabia"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,18]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1145\/3626772.3657744","article-title":"\u201cGraded relevance scoring of written essays with dense retrieval,\u201d","author":"Albatarni","year":"2024","journal-title":"Proceedings of the 47th international ACM SIGIR conference on research and development in information retrieval"},{"key":"B2","first-page":"715","article-title":"\u201cAutomatic text scoring using neural networks,\u201d","volume-title":"Proceedings of the 54th annual meeting of the association for computational linguistics (ACL)","author":"Alikaniotis","year":"2016"},{"key":"B3","doi-asserted-by":"publisher","first-page":"58746","DOI":"10.1109\/ACCESS.2025.3556352","article-title":"ET-GNN: ensemble transformer-based graph neural networks for holistic automated essay scoring","volume":"13","author":"Aljuaid","year":"2025","journal-title":"IEEE Access"},{"key":"B4","first-page":"1650","article-title":"Automated essay scoring with e-rater\u00ae V. 2","volume":"4","author":"Attali","year":"2006","journal-title":"J. Technol. Learn. Assess"},{"key":"B5","doi-asserted-by":"publisher","first-page":"858","DOI":"10.5753\/jbcs.2025.5817","article-title":"An empirical analysis of large language models for automated cross-prompt essay trait scoring in Brazilian Portuguese","volume":"31","author":"Barbosa","year":"2025","journal-title":"J. Brazil. Comput. Soc."},{"key":"B6","article-title":"On the opportunities and risks of foundation models","author":"Bommasani","year":"2021","journal-title":"arXiv preprint arXiv:2108.07258"},{"key":"B7","first-page":"1877","article-title":"\u201cLanguage models are few-shot learners,\u201d","author":"Brown","year":"2020","journal-title":"Advances in neural information processing systems (NeurIPS)"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.884","article-title":"\u201cMixture of ordered scoring experts for cross-prompt essay trait scoring,\u201d","author":"Chen","year":"2025","journal-title":"Proceedings of the 63rd annual meeting of the association for computational linguistics (Volume 1: Long Papers)"},{"key":"B9","first-page":"276","article-title":"A multi-task automated assessment system for essay scoring,\u201d","volume-title":"Proceedings of the international conference on artificial intelligence in education","author":"Chen","year":"2024"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.83","article-title":"\u201cPMAES: prompt-mapping contrastive learning for cross-prompt automated essay scoring,\u201d","author":"Chen","year":"2023","journal-title":"Proceedings of the 61st annual meeting of the association for computational linguistics (Volume 1: Long Papers)"},{"key":"B11","doi-asserted-by":"publisher","first-page":"82","DOI":"10.4218\/etrij.2023-0324","article-title":"Dual\u2013scale BERT using multi\u2013trait representations for holistic and trait\u2013specific essay grading","volume":"46","author":"Cho","year":"2024","journal-title":"ETRI J."},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-naacl.322","article-title":"\u201cRationale behind essay scores: enhancing S-LLM's multi-trait essay scoring with rationale generated by LLMs,\u201d","author":"Chu","year":"2025","journal-title":"Findings of the association for computational linguistics: NAACL 2025"},{"key":"B13","first-page":"4171","article-title":"\u201cBERT: pre-training of deep bidirectional transformers for language understanding,\u201d","volume-title":"Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics (NAACL)","author":"Devlin","year":"2019"},{"key":"B14","first-page":"1640","article-title":"An overview of automated scoring of essays","volume":"5","author":"Dikli","year":"2006","journal-title":"J. Technol. Learn. Assess."},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-acl.98","article-title":"\u201cPrompt- and trait relation-aware cross-prompt essay trait scoring,\u201d","author":"Do","year":"2023","journal-title":"Findings of the association for computational linguistics: ACL 2023"},{"key":"B16","first-page":"153","article-title":"\u201cAttention-based recurrent convolutional neural network for automatic essay scoring,\u201d","volume-title":"Proceedings of the 21st conference on computational natural language learning (CoNLL)","author":"Dong","year":"2017"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1109\/AICT67988.2025.11268740","article-title":"\u201cOptimizing LLMs for automated essay scoring: finetuning and systematic prompting,\u201d","author":"ElMassry","year":"2025","journal-title":"2025 IEEE 19th international conference on application of information and communication technologies (AICT)"},{"key":"B18","doi-asserted-by":"publisher","first-page":"38","DOI":"10.14569\/IJACSA.2020.0110538","article-title":"A trait-based deep learning automated essay scoring system with adaptive feedback","volume":"11","author":"Hussein","year":"2020","journal-title":"Int. J. Adv. Comput. Sci. Applic."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-99264-3_13","article-title":"\u201cAutomated essay scoring using logical structure graph information,\u201d","author":"Kato","year":"2025","journal-title":"Posters and late breaking results, workshops and tutorials, industry and innovation tracks, practitioners, doctoral consortium, blue sky, and WideAIED"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.63317\/45fqsdi69jq5","article-title":"\u201cCan large language models automatically score proficiency of written essays,\u201d","author":"Mansour","year":"2024","journal-title":"Proceedings of the 2024 joint international conference on computational linguistics, language resources and evaluation (LREC-COLING 2024)"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.63317\/4pmfncubhk6c","article-title":"\u201cASAP++: enriching the ASAP automated essay grading dataset with essay attribute scores,\u201d","author":"Mathias","year":"2018","journal-title":"Proceedings of the eleventh international conference on language resources and evaluation (LREC 2018)"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.bea-1.8","article-title":"\u201cCan neural networks automatically score essay traits,\u201d","author":"Mathias","year":"2020","journal-title":"Proceedings of the fifteenth workshop on innovative use of NLP for building educational applications"},{"key":"B23","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1007\/s10462-024-11017-5","article-title":"A survey on deep learning-based automated essay scoring and feedback generation","volume":"58","author":"Misgna","year":"2024","journal-title":"Artif. Intell. Rev."},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1109\/IC2NC67409.2025.11376410","article-title":"\u201cAutomatic English essay grading using convolutional neural networks and semantic similarity scoring,\u201d","author":"Mohsin","year":"2025","journal-title":"2025 international conference on NexGen networks and cybernetics (IC2NC)"},{"key":"B25","article-title":"GPT-4 technical report","year":"2023","journal-title":"arXiv preprint arXiv:2303.08774"},{"key":"B26","first-page":"238","article-title":"The imminence of\u2026 grading essays by computer","volume":"47","author":"Page","year":"1966","journal-title":"Phi Delta Kappan"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1109\/AINIT65432.2025.11035868","article-title":"\u201cMulti-perspective coherence assessment method for English texts,\u201d","author":"Pan","year":"2025","journal-title":"2025 IEEE 6th international seminar on artificial intelligence, networking and information technology (AINIT)"},{"key":"B28","article-title":"Prompt agnostic essay scorer: a domain generalization approach to cross-prompt automated essay scoring","author":"Ridley","year":"2020","journal-title":"arXiv preprint arXiv:2008.01441"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i15.17620","article-title":"\u201cAutomated cross-prompt scoring of essay traits,\u201d","author":"Ridley","year":"2021","journal-title":"Proceedings of the AAAI conference on artificial intelligence"},{"key":"B30","first-page":"1668","article-title":"Automated essay scoring using Bayes' theorem","volume":"1","author":"Rudner","year":"2002","journal-title":"J. Technol. Learn. Assess."},{"key":"B31","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/j.asw.2013.04.001","article-title":"State-of-the-art automated essay scoring: competition, results, and future directions from a United States demonstration","volume":"20","author":"Shermis","year":"2014","journal-title":"Assess. Writ."},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.4324\/9780203122761","author":"Shermis","year":"2013","journal-title":"Handbook of Automated Essay Evaluation"},{"key":"B33","doi-asserted-by":"publisher","first-page":"169","DOI":"10.1007\/978-3-031-79032-4_12","article-title":"\u201cInvestigating universal adversarial attacks against transformers-based automatic essay scoring systems,\u201d","author":"Silveira","year":"2025","journal-title":"Brazilian conference on intelligent systems"},{"key":"B34","doi-asserted-by":"publisher","first-page":"5390","DOI":"10.1007\/s11227-023-05640-2","article-title":"Enhanced cross-prompt trait scoring via syntactic feature fusion and contrastive learning","volume":"80","author":"Sun","year":"2023","journal-title":"J. Supercomput"},{"key":"B35","doi-asserted-by":"crossref","first-page":"1882","DOI":"10.18653\/v1\/D16-1193","article-title":"\u201cA neural approach to automated essay scoring,\u201d","volume-title":"Proceedings of the 2016 conference on empirical methods in natural language processing (EMNLP)","author":"Taghipour","year":"2016"},{"key":"B36","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v32i1.12045","article-title":"\u201cSkipFlow: incorporating neural coherence features for end-to-end automatic text scoring,\u201d","author":"Tay","year":"2018","journal-title":"Proceedings of the AAAI conference on artificial intelligence"},{"key":"B37","doi-asserted-by":"publisher","first-page":"459","DOI":"10.1007\/s41237-021-00142-y","article-title":"A review of deep-neural automated essay scoring models","volume":"48","author":"Uto","year":"2021","journal-title":"Behaviormetrika"},{"key":"B38","doi-asserted-by":"publisher","first-page":"109308","DOI":"10.1016\/j.compeleceng.2024.109308","article-title":"A multifaceted architecture to Automate Essay Scoring for assessing english article writing: integrating semantic, thematic, and linguistic representations","volume":"118","author":"Wang","year":"2024","journal-title":"Comput. Electr. Eng."},{"key":"B39","doi-asserted-by":"crossref","first-page":"24824","DOI":"10.52202\/068431-1800","article-title":"\u201cChain-of-thought prompting elicits reasoning in large language models,\u201d","author":"Wei","year":"2022","journal-title":"Advances in neural information processing systems (NeurIPS)"},{"key":"B40","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511732997","volume-title":"Assessing Writing","author":"Weigle","year":"2002"},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.1109\/COMNETSAT63286.2024.10862459","article-title":"\u201cAtomic evaluation using large language model for automated essay exam scoring,\u201d","author":"Winarta","year":"2024","journal-title":"2024 IEEE international conference on communication, networks and satellite (COMNETSAT)"},{"key":"B42","doi-asserted-by":"publisher","first-page":"129283","DOI":"10.1016\/j.neucom.2024.129283","article-title":"EPCTS: enhanced prompt-aware cross-prompt essay trait scoring","volume":"621","author":"Xu","year":"2025","journal-title":"Neurocomputing"},{"key":"B43","article-title":"Adversarial topic-aware prompt-tuning for cross-topic automated essay scoring","author":"Zhang","year":"","journal-title":"arXiv preprint arXiv:2508.05987"},{"key":"B44","first-page":"1190","article-title":"Category adversarial joint learning method for cross-prompt automated essay scoring","volume":"62","author":"Zhang","year":"","journal-title":"J. Comput. Res. Dev."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1895359\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T05:36:09Z","timestamp":1789709769000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1895359\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,18]]},"references-count":44,"alternative-id":["10.3389\/frai.2026.1895359"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1895359","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,18]]},"article-number":"1895359"}}