Research
I shape AI to help students develop critical thinking through argumentation in STEM, with particular attention to classroom discourse and critical evaluation of AI capabilities.
The throughline: ArguAgent exists to catch small-group arguments that need help before they stall. Building it surfaced an audio-capture problem serious enough to become its own project, VoiceLens. The SSMA and AMTE work below asks the harder question underneath both — when, exactly, does a small group need an intervention?
Active Projects
Argumentation in the classroom
ArguAgent
Platform for supporting mathematical argumentation in collaborative small-group settings. Uses LLMs to analyze student positioning responses and form optimal discussion groups.
Key Innovation
Separates stance identification from correctness evaluation. Groups students by argumentation sophistication (Osborne levels 0-5) rather than right/wrong answers.
Status
Presented at AIED 2026 (paper #1127, Springer LNCS). A second paper from the broader collaboration, on LLM agent communication strategies, was presented at AIED 2025. Pilot testing surfaced a classroom audio-capture problem serious enough to become its own project (VoiceLens). Active development continues.
VoiceLens
Privacy-aware system co-designed with teachers and guardians for capturing small-group argumentation indicators in high-school science classrooms, without recording or retaining raw student audio.
Funding
NSF FINDERS Foundry planning grant ($49,999). PI: Dr. Xiaoming Zhai. Co-PI: Christina Chhin. Jennifer is technologist and researcher on the project, alongside Arne Bewersdorff.
Status
Grew out of an audio-capture problem discovered during ArguAgent's April 2026 pilot testing. IRB approved July 2026; recruitment of Georgia science teachers and their students underway.
Scrutinizing AI as a research tool
INTRA
Systematic methodology for using Gemini's multimodal capabilities to transcribe classroom videos with speaker identification and high reliability validation. Formerly called the Video Transcription Pipeline.
Pipeline
Python-based system using Gemini video API with 2-minute chunks, VAD preprocessing, and RTF-compatible output for Transana import.
Validation
Validated with 3 coders. Confidence markers only appear when quality thresholds aren't met (<70% speaker, <85% text). Currently the subject of a verification study with AnnaMarie Conner, Anna Bloodworth, Joseph Appiah, and Susan Cannon.
Publications
Published
Articulating Human Engagement: Disclosure as Key to Ethical Use of AI in Mathematics Education Research
Horn, I., & Kleiman, J.
Journal for Research in Mathematics Education, 57(4), 148-156. https://doi.org/10.5951/jresematheduc-2026-0062
Editorial from the Math Education Journal Editors AI working group.
ArguAgent: AI-Supported Real-Time Grouping for Productive Argumentation in STEM Classrooms
Kleiman, J., Gao, Y., Xia, X., Wang, Z., Zhu, Z., Park, J., & Zhai, X.
In International Conference on Artificial Intelligence in Education (pp. 298-305). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-032-29760-0_33
Beyond Excel: Leveraging AI for Student Data Exploration
Kleiman, J., & Fitzgerald, K.
Ohio Journal of School Mathematics 101(1):52-70. doi: 10.18061/ojsm.6627
Introduces AILIVE framework for AI-enhanced data visualization literacy in mathematics classrooms.
Expert Analysis of AI-Generated Theoretical Analyses in Mathematics Education
Kleiman, J., Byerley, C., & Moore, K.
PME-NA 2025 Proceedings, pp. 1265–1273. The Pennsylvania State University.
MKT-AI framework: adapting Mathematical Knowledge for Teaching to evaluate AI-generated educational content. Full citation: Kleiman, J., Byerley, C., & Moore, K. C. (2025). Expert analysis of AI-generated theoretical analyses in mathematics education. In R. M. Zbiek, X. Yao, A. McCloskey, & F. Arbaugh (Eds.), Proceedings of the 47th annual meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education (pp. 1265-1273). The Pennsylvania State University. https://www.pmena.org/pmenaproceedings/PMENA%2047%202025%20Proceedings.pdf
Exploring Communication Strategies for Collaborative LLM Agents in Mathematical Problem-Solving
Zhang, L., Zhai, X., Lin, J., Kleiman, J., Zapata-Rivera, D., Forsyth, C., Jiang, Y., Hu, X., & Graesser, A. C.
In International Conference on Artificial Intelligence in Education (pp. 258-265). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-99264-3_32
The Development of Students' Understanding of Permutations and Symmetries
Asiala, M., Kleiman, J., Brown, A. et al.
International Journal of Computers for Mathematical Learning 3, 13–43 (1998). https://doi.org/10.1023/A:1009738109343
Under review
Supporting Small-Group Mathematical Argumentation with Real-Time AI Assistance in the Classroom
Kleiman, J., Gao, Y., Dollarhide, K., Wong, W., & Zhai, X.
Invited chapter for a Springer monograph on AI in mathematics education (Submitted October 2026)
Sunny Summaries by AI: Insidious Textual Productions and the Value of Field Notes
Cannon, S., & Kleiman, J.
International Journal of Social Research Methodology — special issue, "The Complicated and Complicating Role of AI in Qualitative Inquiry" (Submitted April 2026)
Journal counterpart to the AERA 2027 submission "Compulsively Smoothed," from the same collaboration with Susan Cannon.
Presentations
Presented
Can ChatGPT Help Analyze Collaborative Mathematical Argumentation? Comparing AI Tools for Classroom Argumentation Analysis
Kleiman, J., Gao, Y., & Zhai, X.
AERA Annual Meeting, Los Angeles, CA [Interactive presentation]
Teacher Professional Development on AI Integration: A Qualitative Study of Programs and Pedagogies
Ding, A.-C. E., Kleiman, J., Zhang, Y., Shi, L., Zhang, K., Jiang, Z., & Zhai, X.
AERA Annual Meeting, Los Angeles, CA [Poster 53]
How Can We Integrate Mathematics and Coding in Elementary Classrooms?
Wong, W., Bloodworth, A., Kleiman, J., & Conner, A.
AMTE Annual Conference, Portland, OR [Conference session]
Beyond Excel: Leveraging AI for Student Data Exploration in Mathematics Classrooms
Kleiman, J., & Fitzgerald, K.
Georgia Mathematics Conference 2025
Empowering Students' Argumentation Skills Through AI: A Healthy Adversarial Approach
Kleiman, J.
NCTM 2025 Atlanta Annual Meeting [Poster]
About Build-A-Bot, a system since discontinued.
Using AI to Do Qualitative Video Research in Mathematics Education
Kleiman, J., Zhai, X., & Ellis, A.
AERA Annual Meeting, Denver, CO
Qualitative Research Data Analysis with an AI Co-Creator: Participative Epistemology in Action
Kleiman, J.
Invited colloquium, Georgia Southern University (virtual)
Using Google NotebookLM for Academic Research
Kleiman, J.
Technology Exploration in Action (TEA) seminar series
NotebookLM and Its Implications for Research and Academic Writing
Kleiman, J.
TEA editor meeting [Discussion leader]
AI in Research Literature Engagement
Kleiman, J.
Invited presentation
AI Techniques to Find and Manage Research Literature: Making Haystacks, Finding Needles
Kleiman, J.
Invited colloquium, Georgia Southern University (virtual)
AI Like a Boss: Georgia Southern Edition
Kleiman, J.
Invited colloquium hosted by Dr. Heidi Eisenreich, Georgia Southern University, Statesboro, GA
AI Like a Boss: Learn the How-Tos
Kleiman, J., & Wilbanks, L.
Georgia Mathematics Conference 2024 [Invited presentation]
Fostering Strong Critical Thinking and Healthy Adversarial Relationships with Technology
Kleiman, J., & Fitzgerald, K.
GAMTE 2024 [Invited presentation]
AI Literacy and Ethical Practice in Education
Kleiman, J.
Invited speaker, Fall 2024 ITT Academy, University of Georgia
Generative AI in Science Education
Kleiman, J.
Invited speaker, Science Education Doctoral Seminar, University of Georgia
Accepted
Sensing Intervention Opportunities in Small-Group Mathematical Argumentation
Kleiman, J., Tembe, N., & Conner, A.
125th School Science and Mathematics Association Convention, Chicago, IL [Research Session Presentation]
Accompanying manuscript accepted to the SSMA proceedings volume. Draws on data from the C4OMS project (AnnaMarie Conner, PI).
Robot Delivery: A Design Challenge for Integrated STEM and Supporting Collective Argumentation
Bloodworth, A., Wong, W., Tembe, N., Kleiman, J., & Conner, A.
125th School Science and Mathematics Association Convention, Chicago, IL [Workshop]
Conceptualizing Argument(ation) Quality in Mathematics and STEM Classrooms
Foster, C., Kleiman, J., Conner, A., Singletary, L., & Bieda, K.
AMTE 2026 Virtual Sessions [Symposium]
ArguAgent: Voice-Based AI Support for Enhancing Development of Mathematical Argumentation Through Small Group Discussions
Kleiman, J.
JMM 2027, Chicago, IL, January 12, 2027 — AMS-MAA Special Session on Engaging Mathematics Learners with AI II: New Frontiers in Teaching, Learning, and Research, II
What Is an Intervention Opportunity in Small-Group Mathematical Argumentation?
Kleiman, J., Tembe, N., & Conner, A.
31st Annual AMTE Conference, Baltimore, MD, February 11-13, 2027
Under review
The Prompt as Apparatus: Multimodal AI Transcription for Classroom Research
Kleiman, J., Appiah, J., Bloodworth, A., & Cannon, S.
AERA 2027 Annual Meeting (Submitted July 2026)
Two Ways a Transcript Fails: Microphone Placement as a Validity Threat in Small-Group Research
Kleiman, J., Bewersdorff, A., & Zhai, X.
AERA 2027 Annual Meeting (Submitted July 2026)
Compulsively Smoothed: Insidious AI Summaries and the Value of Field Notes
Cannon, S., & Kleiman, J.
AERA 2027 Annual Meeting (Submitted July 2026)
Grants & Funding
NSF FINDERS Foundry Planning Grant
NSF FINDERS Foundry planning grant ($49,999). PI: Dr. Xiaoming Zhai. Co-PI: Christina Chhin. Jennifer is technologist and researcher on the project, alongside Arne Bewersdorff.
Mathematics Education Summer Research Scholarship
$4,000, University of Georgia, Summer 2025. Faculty Advisor: Dr. AnnaMarie Conner.
The Mathematics Educator Editorial Support Funding
$1,000, University of Georgia, Summer 2025.
Collaborations
AI PD GENIUS
NSF-funded project developing K-12 AI literacy learning companion with Dr. Xiaoming Zhai (PI)
C4OMS
NSF CS for All-funded project ($1M) integrating coding into grades 3-5 math and science, with Dr. AnnaMarie Conner (PI). The "Sensing Intervention Opportunities" paper draws on data from this project.
Service
Editor, The Mathematics Educator
Co-editing the journal since Fall 2024, alongside Sohei Yasuda.