We are delighted to share some memorable moments from the AAAI-25 AI4EDU Workshop! Thank you to all speakers, participants, and organizers for making this event a great success!

Workshop Schedule

Date: March 03, 2025

Location: Pennsylvania Convention Center, Philadelphia, PA, USA

Room: 122A

2:00 PM - 2:10 PM | Opening Remarks

2:10 PM - 2:50 PM | Keynote Talk One + QA

• Speaker: Jill Burstein
• Title: Responsible AI for Leverage Points in Digital Assessment

2:50 PM - 3:30 PM | Keynote Talk Two + QA

• Speaker: Maciej Pankiewicz
• Title: Generative AI in Education: Enhancing Learning, Feedback, and Research with LLMs

3:30 PM - 4:15 PM | Poster Session One

• Workshop Paper Posters:
  • Integrating AI for Mother Tongue Language Proficiency: A Modular Approach
  • SMARTe-VR: Student Monitoring and Adaptive Response Technology for e-learning in Virtual Reality
  • MathVC: An LLM-Simulated Multi-Character Virtual Classroom for Mathematics Education
  • EduBot – Can LLMs Solve Personalized Learning and Programming Assignments?
  • Understanding Disruptions to Virtual Learning: How Much Instructional Time Is Lost?
  • Personalized Paths to Mastery: Real-Time Adaptive Feedback for Student Engagement in Discrete Mathematics
  • Optimizing Multimodal Large Language Models for Scientific VQA through Caption-Aware Supervised Training
• AIED Mini Doctoral Consortium Posters:
  • Accelerating Educational Innovations through Simulations and Cost-Efficient Experimental Design
  • Construction andValidation of an Online Learning Support Model with Individualized Comments

4:15 PM - 5:00 PM | Poster Session Two

• Workshop Paper Posters:
  • Can LLMs identify gaps and errors in students’ code explanations?
  • Enhanced Question-Answering for Skill-based learning using Knowledge-based AI and Generative AI
  • Controllable Text Adaptation Using In-context Learning with Linguistic Features
  • CurriculumAgents: Automated Multi-Agent Lesson Design
  • Functional Near-Infrared Spectroscopy (fNIRS) Analysis of Interaction Techniques in Touchscreen-Based Educational Gaming

Keynote Talk Details

Responsible AI for Leverage Points in Digital Assessment

Jill Burstein

Abstract: As AI-driven educational technologies continue to expand, so do the complexities of digital assessment systems. This keynote explores how a structured ecosystem approach can help identify key leverage points—design, measurement, security, and responsible AI—ensuring validity, fairness, and equity in assessments. By integrating human-centered, responsible AI practices, we can mitigate risks, for instance, to improve test-taker experiences, and uphold rigorous standards in automated scoring, content generation, and security protocols. With global responsible AI guidelines on the rise, this talk will provide a roadmap for researchers and practitioners to embed ethical AI practices into digital assessment frameworks, shaping the future of AI in education.

Bio: Dr. Jill Burstein is a Principal Assessment Scientist at Duolingo, where she leads validity and efficacy research for the Duolingo English Test (DET). She is a co-founder of SIG EDU, a Special Interest Group on Building Educational Applications under the Association for Computational Linguistics (ACL). Her research focuses on artificial intelligence and natural language processing, educational measurement, equity in education, learning analytics, and linguistics. Dr. Burstein pioneered the first automated writing evaluation system used in large-scale, high-stakes assessments, laying the foundation for AI-driven writing assessment. Her recent work explores responsible AI and the validation of digital language assessments. She has published extensively in top academic conferences and journals in the fields of computational linguistics, education, and assessment.

Generative AI in Education: Enhancing Learning, Feedback, and Research with LLMs

Maciej Pankiewicz

Abstract: Generative artificial intelligence is transforming education by improving instruction, automating feedback, and supporting research. This presentation draws on practical experience with Large Language Models (LLMs) in educational settings, from virtual teaching assistants that engage with students on discussion forums to automated assessment tools that provide instant feedback on programming assignments. Beyond instruction, we will examine how LLMs can support qualitative research through automated coding and enhance privacy compliance by de-identifying forum posts. By analyzing the impact of these applications, we will explore both the opportunities and challenges of integrating LLMs into education.

Bio: Dr. Maciej Pankiewicz is a Senior Research Investigator and Visiting Assistant Professor at the University of Pennsylvania’s Graduate School of Education (Penn GSE). He also serves as a researcher at the Penn Center for Learning Analytics. Dr. Pankiewicz earned his Ph.D. in Engineering from the University of Bonn in 2011 and his M.Sc. in Computer Science and Econometrics from the Warsaw University of Life Sciences in 2006. His research interests include educational data mining, learning analytics, and learning engineering. Dr. Pankiewicz leads the development of JeepyTA, an AI-driven virtual teaching assistant powered by large language models, aimed at enhancing educational experiences for both students and instructors. Dr. Pankiewicz has published extensively in top journals and conferences. His research has been recognized with multiple nominations for best paper awards at international conferences, including ICQE 2024 and ICCE 2023.  

Accepted Paper List

• Jing Hao Lim, Mohamed Raasith Mohamed Sirajudeen, Justin Kan, Pranav Tushar, Bowen Zhang, Yin Yin Loo, Indriyati Atmosukarto, Donny Soh and Ian McLoughlin. Integrating AI for Mother Tongue Language Proficiency: A Modular Approach.

• Roberto Daza, Lin Shengkai, Aythami Morales, Julian Fierrez and Katashi Nagao. SMARTe-VR: Student Monitoring and Adaptive Response Technology for e-learning in Virtual Reality. Paper Link

• Maryam Ebrahimi, Rajeev Sahay, Seyyedali Hosseinalipour and Bita Akram. Vision Paper: Advancing Mental Health in Education through Federated Learning and its Synergies with Broader Human-Centered Domains. Paper Link

• Lucile Favero, Juan Antonio Pérez-Ortiz, Tanja Kaser and Nuria Oliver. Argument Mining in Education: Exploring the Potential of Open-source Small LLMs for Argument Classification and Assessment. Paper Link

• Murong Yue, Wenhan Lyu, Wijdane Mifdal, Jennifer Suh, Yixuan Zhang and Ziyu Yao. MathVC: An LLM-Simulated Multi-Character Virtual Classroom for Mathematics Education. Paper Link

• Yibin Wang, Jiaxi Xie and Lakshminarayanan Subramanian. EduBot – Can LLMs Solve Personalized Learning and Programming Assignments? Paper Link

• Likhitha Paruchuri, Emma Terrell, Monica Figueroa, Nonye Alozie and Anirban Roy. A Multimodal Framework for Identifying Science Concepts in Educational Videos. Paper Link

• Ayan, Mukesh Mohania and Vikram Goyal. ASAGLite: A Lightweight and Efficient Technique for Automatic Short Answer Grading. Paper Link

• Xander Beberman, Sarah Novicoff, Ana Trindade Ribeiro, Carly Robinson and Susanna Loeb. Understanding Disruptions to Virtual Learning: How Much Instructional Time Is Lost? Paper Link

• Anthony Cusimano and Renzhe Yu. Mapping Longitudinal Pathways of Learning Strategies with Fair Multi-Level Clustering of Clickstream Data. Paper Link

• Andre Kenneth Chase Randall, Jasmine V. Ngo, Beverly P. Woolf and Andrew Lan. Personalized Paths to Mastery: Real-Time Adaptive Feedback for Student Engagement in Discrete Mathematics. Paper Link

• Janak Kapuriya, Arnav Goel, Medha Hira, Apoorv Singh, Naman Lal, Jay Saraf, Sanjana, Vaibhav Nautiyal, Avinash Anand and Rajiv Ratn Shah. Optimizing Multimodal Large Language Models for Scientific VQA through Caption-Aware Supervised Training. Paper Link

• Priti Oli, Rabin Banjade, Andrew M. Olney and Vasile Rus. Can LLMs identify gaps and errors in students’ code explanations? Paper Link

• Rahul K. Dass, Rochan H. Madhusudhana, Erin C. Deye, Shashank Verma, Timothy A. Bydlon, Grace Brazil and Ashok K. Goel. Enhanced Question-Answering for Skill-based learning using Knowledge-based AI and Generative AI. Paper Link

• Zheng Guan, Jie Liu, Zhijun Yang and Xue Wang. A Dataset and Cross-subject Enhanced Model for Cognitive Diagnosis. Paper Link

• Sarubi Thillainathan and Alexander Koller. Controllable Text Adaptation Using In-context Learning with Linguistic Features. Paper Link

• Edward Sun, Yijia Xiao and Wei Wang. CurriculumAgents: Automated Multi-Agent Lesson Design. Paper Link

• Chao Zhang, Jiamin Tang and Jing Xiao. Tangram: Benchmark for Evaluating Geometric Element Recognition in Large Multimodal Models. Paper Link

• Shayla Sharmin, Elham Bakhshipour, Mohammad Fahim Abrar, Behdokht Kiafar, Pinar Kullu, Nancy Getchell and Roghayeh Leila Barmaki. Functional Near-Infrared Spectroscopy (fNIRS) Analysis of Interaction Techniques in Touchscreen-Based Educational Gaming. Paper Link

Introduction

Motivation

The rapid advancement of generative AI technologies presents both unprecedented opportunities and significant challenges within the educational landscape. With tools like ChatGPT, DALL-E, and other generative models becoming increasingly sophisticated, educators have access to powerful resources that can enhance learning experiences, personalize education, and streamline administrative tasks. However, these advancements also bring forth critical issues such as ethical considerations, data privacy concerns, and potential biases. This workshop aims to explore how generative AI can be effectively and responsibly integrated into education, ensuring that its benefits are maximized while mitigating associated risks.

Challenges

• Ethical Considerations: Ensuring that generative AI is used ethically in educational settings, addressing issues related to bias, fairness, and transparency. Addressing inherent biases in AI algorithms that could lead to unequal learning opportunities or reinforce existing disparities.
• Data Privacy: Protecting student data and ensuring compliance with privacy regulations when using AI tools.
• Integration and Adoption: Overcoming barriers to integrating AI tools into traditional educational systems, including resistance from educators and lack of technical infrastructure.
• Skill Gaps: Equipping educators with the necessary skills and knowledge to effectively use AI tools in their teaching practices.
• Evaluation and Effectiveness: Developing robust methods to assess the effectiveness of AI-driven educational tools and their impact on learning outcomes.

Goals

• Knowledge Sharing and Collaboration: To provide a platform for researchers, educators, and industry professionals to share insights, experiences, and best practices regarding the use of generative AI in education. To foster a collaborative environment where participants can engage in discussions, exchange ideas, and form partnerships aimed at advancing the responsible use of AI in education.
• Exploration of Tools: To showcase innovative AI tools and applications that can transform educational practices, highlighting both their capabilities and limitations.
• Risk Mitigation Strategies: To discuss strategies for identifying and mitigating the risks associated with generative AI, focusing on ethical use, data privacy, and bias reduction.
• Policy Recommendations: To develop guidelines and recommendations for policymakers and educational institutions on the adoption and regulation of generative AI technologies.
• Future Directions: To identify emerging trends and future research directions in the field of AI for education, encouraging continued innovation and development.

By addressing these motivations, challenges, and goals, this workshop seeks to contribute significantly to the discourse on generative AI in education, promoting its responsible and effective integration to improve learning outcomes and educational equity.

Workshop Description

In this workshop, we will invite AIED enthusiasts from all around the world through the following three different channels:

• First, we will invite established researchers in the AIED community to give a keynote talk that (1) describes a vision for bridging AIED communities; (2) summarizes a well-developed AIED research area; or (3) presents promising ideas and visions for new AIED research directions.

• Second, we will call for regular workshop paper submissions related to a broad range of AI domains for education.

• Third, we will host an AIED doctoral consortium that provides an opportunity for a group of Ph.D. students to discuss and explore their research interests and career objectives with a panel of established AIED researchers.

Through these initiatives, we aim to provide a common ground for researchers to share their cutting-edge insights on AIED and encourage the development of practical and large-scale AIED methods of lasting impact.

Paper Submissions

We invite high-quality paper submissions of a theoretical and experimental nature on generative AI topics including, but not limited to, the following:

• Emerging technologies in education
• Evaluation of education technologies
• Immersive learning and multimedia applications
• Self-adaptive learning
• Individual and personalized education
• Intelligent learning systems
• Intelligent tutoring and monitoring systems
• Automatic grading and assessment
• Automated feedback and recommendations
• Big data analytics for education
• Analysis of communities of learning
• Course development techniques
• Data analytics & big data in education
• Mining and web mining in education
• Learning tools experiences and cases of study
• Social media in education
• Smart education
• Digital libraries for learning
• Knowledge management for learning
• Learning technology for lifelong learning
• Tracking learning activities
• Wearable computing technology in e-learning
• Smart classroom
• Dropout prediction
• Knowledge tracing

The workshop solicits 5-7 pages double-blind paper submissions (with unlimited references) from participants. Submissions of the following flavors will be sought: (1) research ideas, (2) case studies (or deployed projects), (3) review papers, (4) best practice papers, and (5) lessons learned. The format is the standard double-column AAAI Proceedings Style. All submissions will be peer-reviewed. Some will be selected for spotlight talks, and some for the poster session.

Papers can be submitted through the OpenReview link: https://openreview.net/group?id=AAAI.org/2025/Workshop/AI4EDU

AIED Doctoral Consortium

The AIED Doctoral Consortium for PhD students researching AI in Education aims to create a focused and supportive environment where emerging scholars can engage with experienced researchers. The primary goals of this event are: (1) Research Feedback and Guidance: To offer participants constructive feedback on their current research projects and advice on future directions. (2) Community Building: To foster a collaborative and supportive community among doctoral students and established researchers. (3) Career Development: To provide insights into various career paths, including academic, industry, and nontraditional roles, thus supporting the professional growth of diverse researchers. (4) Conference Integration: To enhance the overall conference experience by facilitating interactions between consortium attendees and other conference participants.

This specialized consortium is designed to help PhD students refine their research, build networks, and navigate their career paths within the niche field of AI in Education.

Detailed about the AIED Doctoral Consortium can be found at https://ai4ed.cc/workshops/aaai2025consortium.

Important Dates

Note: All deadlines are anywhere on earth (UTC-12)

• November 22 December 15, 2024: Workshop Submissions Due
• December 9 December 25, 2024: Notifications Sent to Authors
• December 19, 2024: AAAI-25 Early Registration Deadline
• March 3, 2025: AAAI-25 AI4EDU Workshop Program

Organizers

• Zitao Liu Guangdong Institute of Smart Education, Jinan University, China
• John Stamper Human-Computer Interaction Institute, Carnegie Mellon University, USA
• Andrew M. Olney Department of Psychology, University of Memphis, USA
• Tianqiao Liu TAL Education Group, China
• Qingsong Wen Squirrel AI Learning, USA
• Jiliang Tang Michigan State University, USA
• Joleen Liang Squirrel AI Learning, USA