First Call for Papers

The 21st Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2026) San Diego, California, United States and online Thursday, July 2 and Friday, July 3, 2026 (co-located with ACL 2026) https://sig-edu.org/bea/current Submission Deadline: Thursday, March 5, 2026, 11:59pm UTC-12

WORKSHOP DESCRIPTION The BEA Workshop is a leading venue for NLP innovation in the context of educational applications. It is one of the largest one-day workshops in the ACL community with over 100 registered attendees in the past several years. The growing interest in educational applications and a diverse community of researchers involved resulted in the creation of the Special Interest Group in Educational Applications (SIGEDU)https://sig-edu.org/members in 2017, which currently has over 400 members.

The 21st BEA will be a 2-day workshop, with one in-person workshop day and one virtual workshop day. The workshop will feature oral presentation sessions and large poster sessions to facilitate the presentation of a wide array of original research. Moreover, there will be a panel discussion on “Transitioning from Academia to the EdTech Industry”, a half-day tutorial on Theory of Mind and its Applications in Educational Contexts, and two shared tasks on Vocabulary Difficulty Prediction for English Learnershttps://www.britishcouncil.org/data-science-and-insights/bea2026st and on Rubric-based Short Answer Scoring for Germanhttps://edutec.science/bea-2026-shared-task/ comprising an oral overview presentation by the shared task organizers and several poster presentations by the shared task participants.

The workshop will accept submissions of both full papers and short papers, eligible for either oral or poster presentation. We solicit papers that incorporate NLP methods, including, but not limited to:

* use of generative AI in education and its impact; * automated scoring of open-ended textual and spoken responses; * automated scoring/evaluation for written student responses (across multiple genres); * game-based instruction and assessment; * educational data mining; * intelligent tutoring; * collaborative learning environments; * peer review; * grammatical error detection and correction; * learner cognition; * spoken dialog; * multimodal applications; * annotation standards and schemas; * tools and applications for classroom teachers, learners and/or test developers; and * use of corpora in educational tools.

SHARED TASKS

Vocabulary Difficulty Prediction for English Learners Organizers: Mariano Felice (British Council) and Lucy Skidmore (British Council). Description: This shared task aims to advance research into vocabulary difficulty prediction for learners of English with diverse L1 backgrounds, an essential step towards custom content creation, computer-adaptive testing and personalised learning. In a context where traditional item calibration methods have become a bottleneck for the implementation of digital learning and assessment systems, we believe predictive NLP models can provide a more scalable, cost-effective solution. The goal of this shared task is to build regression models to predict the difficulty of English words given a learner’s L1. We believe this new shared task provides a novel approach to vocabulary modelling, offering a multidimensional perspective that has not been explored in previous work. To this aim, we will use the British Council’s Knowledge-based Vocabulary Lists (KVL), a multilingual dataset with psychometrically calibrated difficulty scores. We believe this unique dataset is not only an invaluable contribution to the NLP community but also a powerful resource that will enable in-depth investigations into how linguistic features, L1 background and contextual cues influence vocabulary difficulty.

For more information on how to participate and latest updates, please refer to the shared task website: https://www.britishcouncil.org/data-science-and-insights/bea2026st

Rubric-based Short Answer Scoring for German Organizers: Sebastian Gombert (DIPF), Zhifan Sun (DIPF), Fabian Zehner (DIPF), Jannik Lossjew (IPN), Tobias Wyrwich (IPN), Berrit Katharina Czinczel (IPN), David Bednorz (IPN), Sascha Bernholt (IPN), Knut Neumann (IPN), Ute Harms (IPN), Aiso Heinze (IPN), and Hendrik Drachsler (DIPF)

Description: Short answer scoring is a well-established task in educational natural language processing. In this shared task, we introduce and focus on rubric-based short-answer scoring, a task formulation in which models are provided with a question, a student answer, and a textual scoring rubric that specifies criteria for each possible score level. Successfully solving this task requires models to interpret the semantics of scoring rubrics and apply their criteria to previously unseen answers, closely mirroring how human raters assign scores in educational assessment. Although rubrics have been used as auxiliary information in prior work on free-text scoring and LLM-based approaches, there has been little focused investigation of rubric-based short-answer scoring as a task in its own right. This setting poses distinct challenges, including ambiguous or underspecified rubric criteria and a wide range of valid student responses. With this shared task, we aim to stimulate systematic research on rubric-based scoring, assess how well current NLP methods can reason over rubrics, and identify promising modeling strategies. Additionally, by providing a German-language dataset, the shared task contributes a new non-English benchmark to the field.

For more information on how to participate and latest updates, please refer to the shared task website: https://edutec.science/bea-2026-shared-task/

TUTORIAL

Theory of Mind and Application in Educational Context Organizers: Effat Farhana (Auburn University), Maha Zainab (Auburn University), Qiaosi Wang (Carnegie Mellon University), Niloofar Mireshghallah (Carnegie Mellon University), Ramira van der Meulen (Leiden University), Max van Duijn (Leiden University). Description: This tutorial examines the integration of Theory of Mind (ToM) into AI-driven online tutoring systems, focusing on how advanced technologies, such as Large Language Models (LLMs), can model learners’ cognitive and emotional states to provide adaptive, personalized feedback. Participants will learn foundational principles of ToM from cognitive science and psychology and how these concepts can be operationalized in AI systems. We will discuss mutual ToM, where both AI tutors and learners maintain models of each other’s mental states, and address challenges such as detecting learner misconceptions, modeling meta-cognition, and maintaining privacy in data-driven tutoring. The tutorial also presents hands-on demonstrations of Machine ToM applied to programming education using datasets such as CS1QA and CodeQA, which contain Java and Python samples. By combining conceptual foundations, research insights, and practical exercises, this tutorial provides a comprehensive overview of designing human-centered, ethically aware, and cognitively informed AI tutoring systems.

IMPORTANT DATES All deadlines are 11.59 pm UTC-12 (anywhere on earth).

* Submission deadline: Thursday, March 5, 2026 * Notification of acceptance: Tuesday, April 28, 2026 * Camera-ready papers due: Tuesday, May 12, 2026 * Workshop: Thursday, July 2, and Friday, July 3, 2026

SUBMISSION INFORMATION We will be using the ACL Submission Guidelines for the BEA Workshop this year. Authors are invited to submit a long paper of up to eight (8) pages of content, plus unlimited references; final versions of long papers will be given one additional page of content (up to 9 pages) so that reviewers’ comments can be taken into account. We also invite short papers of up to four (4) pages of content, plus unlimited references. Upon acceptance, short papers will be given five (5) content pages in the proceedings. Authors are encouraged to use this additional page to address reviewers’ comments in their final versions. We generally follow ACL submission guidelines and will require that all submitted papers should include a dedicated "Limitations" section, which does not count toward the page limit.

Papers which describe systems are also invited to give a demo of their system. If you would like to present a demo in addition to presenting the paper, please make sure to select either “long paper + demo” or “short paper + demo” under “Submission Category” in the START submission page.

Previously published papers cannot be accepted. The submissions will be reviewed by the program committee. As reviewing will be blind, please ensure that papers are anonymous. Self-references that reveal the author’s identity, e.g., “We previously showed (Smith, 1991) …”, should be avoided. Instead, use citations such as “Smith previously showed (Smith, 1991) …”.

We have also included conflict of interest in the submission form. You should mark all potential reviewers who have been authors on the paper, are from the same research group or institution, or who have seen versions of this paper or discussed it with you.

We will be using the START conference system to manage submissions. The link will be provided soon.

DOUBLE SUBMISSION POLICY We will follow the official ACL double-submission policy. Specifically, papers being submitted both to BEA and another conference or workshop must:

* Note on the title page the other conference or workshop to which they are being submitted. * State on the title page that if the authors choose to present their paper at BEA (assuming it was accepted), then the paper will be withdrawn from other conferences and workshops.

ORGANIZING COMMITTEE

* Ekaterina Kochmar, MBZUAI * Andrea Horbach, Hildesheim University * Ronja Laarmann-Quante, Ruhr University Bochum * Marie Bexte, FernUniversität in Hagen * Anaïs Tack, KU Leuven, imec * Victoria Yaneva, National Board of Medical Examiners * Bashar Alhafni, MBZUAI * Zheng Yuan, University of Sheffield * Jill Burstein, Duolingo * Stefano Bannò, Cambridge University

Workshop contact email address: bea.nlp.workshop@gmail.commailto:bea.nlp.workshop@gmail.com

PROGRAM COMMITTEE Tazin Afrin; David Alfter; Bashar Alhafni; Maaz Amjad; Nischal Ashok Kumar; Stefano Bannò; Michael Gringo Angelo Bayona; Lee Becker; Beata Beigman Klebanov; Luca Benedetto; Bhavya Bhavya; Serge Bibauw; Ted Briscoe; Dominique Brunato; Jie Cao; Dan Carpenter; Jeevan Chapagain; Guanliang Chen; Mei-Hua Chen; Christopher Davis; Orphee De Clercq; Kordula De Kuthy; Jasper Degraeuwe; Dushyanta Dhyani; Yuning Ding; Rahul Divekar; Kosuke Doi; Mohsen Dorodchi; Yo Ehara; Hamza El Alaoui; Sarra El Ayari; Andrew Emerson; Yao-Chung Fan; Mariano Felice; Nigel Fernandez; Michael Flor; Thomas François; Thomas Gaillat; Ananya Ganesh; Ritik Garg; Sebastian Gombert; Samuel González López; Cyril Goutte; Abigail Gurin Schleifer; Na-Rae Han; Ching Nam Hang; Jiangang Hao; Aki Härmä; Hasnain Heickal; Chieh-Yang Huang; Chung-Chi Huang; Radu Tudor Ionescu; Elsayed Issa; N J Karthika; Anisia Katinskaia; Elma Kerz; Fazel Keshtkar; Grandee Lee; Ji-Ung Lee; Arun Balajiee Lekshmi Narayanan; Jiazheng Li; Anastassia Loukina; Wanjing Anya Ma; Jakub Macina; Lieve Macken; Nitin Madnani; Arianna Masciolini; Detmar Meurers; Michael Mohler; Phoebe Mulcaire; Ricardo Muñoz Sánchez; Sungjin Nam; Diane Napolitano; Huy Nguyen; S Jaya Nirmala; Sergiu Nisioi; Michael Noah-Manuel; Adam Nohejl; Amin Omidvar; Daniel Oyeniran; Robert Östling; Ulrike Pado; Yannick Parmentier; Ted Pedersen; Mengyang Qiu; Martí Quixal; Chatrine Qwaider; Arjun Ramesh Rao; Vivi Peggie Rantung; Manikandan Ravikiran; Hanumant Redkar; Robert Reynolds; Saed Rezayi; Frankie Robertson; Aiala Rosá; Andreas Säuberli; Nicy Scaria; Ronald Seoh; Pritam Sil; Astha Singh; Lucy Skidmore; Maja Stahl; Katherine Stasaski; Helmer Strik; Hakyung Sung; Sowmya Vajjala; Elena Volodina; Nikhil Wani; Alistair Willis; Fabian Zehner.