Publications
For a full list, see my Google Scholar.
Human-AI Teaming
How AI teammates enters groups and change group dynamics
Collaboration Analytics
Using ML and NLP to understand and measure how people collaborate
Algorithmic Fairness
Detecting and mitigating bias in educational algorithms, especially for marginalized learners.
Educational Data Science
Using ML and NLP to understand and improve learning outcomes.
🤖 Human-AI Teaming
AI is no longer just a tool now -- it can act as a teammate with sociocognitive and affective capabilities. I study how AI teammates reshape group dynamics and knowledge co-construction process, and design AI teammates that adapt to their teammates' group dynamics to be more socially responsive.
- ArxivMeasuring Inclusion in Interaction: Inclusion Analytics for Human-AI Collaborative LearningarXiv preprint arXiv:2602.09269, 2026
- LAK 2026Read the Room or Lead the Room: Understanding Socio-Cognitive Dynamics in Human-AI TeamingAccepted at International Learning Analytics and Knowledge Conference (LAK), 2026
- CSCL 2026Reconceptualizing Activity Theory for Human-AI Teaming in Computer-Supported Collaborative LearningIn Proceedings of the International Conference of the Computer-Supported Collaborative Learning (CSCL), 2026
đź”— Collaboration Analytics
Collaboration is central to how we learn—and I use machine learning and natural language processing to understand collaborative dynamics. Specifically, I develop Inclusion Analytics, an NLP method to measure how inclusion manifests through participation equity, epistemic dynamics, and sense of belonging in collaborative discourse.
- ArxivMeasuring Inclusion in Interaction: Inclusion Analytics for Human-AI Collaborative LearningarXiv preprint arXiv:2602.09269, 2026
- LAK 2025Understanding Collaborative Learning Processes and Outcomes through Student Discourse DynamicsIn Proceedings of the 15th International Learning Analytics and Knowledge Conference, 2025
- AIED 2025Agentic Men, Communal Women?: Exploring Gender Bias in LLM-Based Leadership Identification for Collaboration AnalyticsIn International Conference on Artificial Intelligence in Education, 2025
⚖️ Algorithmic Fairness
Algorithms in education are not value-neutral. I examine how biases emerge in educational algorithms — especially for minority and marginalized groups — and develop strategies to reliably measure and mitigate them.
- LAK 2025Bias or Insufficient Sample Size? Improving Reliable Estimation of Algorithmic Bias for Minority GroupsIn Proceedings of the 15th International Learning Analytics and Knowledge Conference, 2025
- LAK 2025The Difficulty of Achieving High Precision with Low Base Rates for High-Stakes InterventionIn Proceedings of the 15th International Learning Analytics and Knowledge Conference, 2025
- AIED 2026When Features Misrepresent Underrepresented Learners: Auditing Algorithmic Bias with Differentially Expressive FeaturesIn Proceedings of the International Conference on Artificial Intelligence in Education, 2026
đź“Š Educational Data Science
I use Machine Learning and Natural Language Processing to understand how we learn, and develop computational methods to improve learning outcomes.
- AIED 2024ChatGPT for Education Research: Exploring the Potential of Large Language Models for Qualitative Codebook DevelopmentIn International Conference on Artificial Intelligence in Education, 2024
- ICQEDoes Active Learning Reduce Human Coding?: A Systematic Comparison of Neural Network with nCoderIn International Conference on Quantitative Ethnography, 2022Nominated for Best Paper Award
- PreprintAnchor is the Key: Toward Accessible Automated Essay Scoring with Large Language Models through Prompting2025
- ICLS 2026Redistributing Cognition in AI-Supported WritingIn Proceedings of the International Conference of the Learning Sciences (ICLS), 2026