News

PLUS named a 2026 Tools Competition finalist in the Datasets for Education Innovation track.

For ResearchersPioneering Research in Human-Centered AI for Personalized Math Learning

Our Research Partners

A strategic alliance of world-class universities and industry leaders committed to rigorous learning engineering at scale.

Research Index

Explore the full archive of PLUS research.

Showing 20 of 20 in this preview (newest first)

Tags

Year

Conference

Improving Open-Response Assessment with LearnLM(opens in new tab)
Gen AITutor Training
Demonstrates LearnLM as a more accurate tool than general LLMs for scoring open-ended tutor responses.
2025 · AIED 2025 · Danielle R. Thomas, Conrad Borchers et al.
Intelligent Support for Practice Goal Setting to Enhance Learning(opens in new tab)
Student Learning
This doctoral consortium presentation proposes research on intelligent, data-driven goal-setting to enhance engagement and self-regulation in AI-supported learning. By integrating adaptive feedback and goal recommendations into hybrid tutoring environments, the work aims to advance theory on effort calibration and self-regulated learning while informing the design of scalable, low-overhead supports for active learning.
2025 · 26th International Conference on Artificial Intelligence in Education · Conrad Borchers, Kenneth R. Koedinger et al.
Human Tutoring Improves the Impact of AI Tutor Use on Learning Outcomes(opens in new tab)
Student Learning
Human-AI hybrid tutoring boosts learning more than AI alone; human tutors help students better capitalize on AI tutor time.
2025 · AIED 2025 · Ashish Gurung, Jionghao Lin et al.
Engagement and Learning Benefits of Goal Setting with Rewards in Human-AI Tutoring(opens in new tab)
Student Learning
Weekly practice time rose ~25% and skills mastered per week rose ~40% after goal setting in a 12-week hybrid human-AI tutoring program.
2025 · 26th International Conference on Artificial Intelligence in Education · Conrad Borchers, Alex Houk et al.
VTutor for high-impact tutoring at scale: Managing engagement and real-time multi-screen monitoring with P2P connections.(opens in new tab)
Gen AITutor Training
VTutor is a web-based platform designed to enhance hybrid tutoring by enabling a single tutor to monitor and support multiple students in real time. It combines peer-to-peer screen sharing with AI-powered virtual avatars to detect off-task or struggling students and deliver timely, context-aware feedback.
2025 · In Proceedings of the Twelth ACM Conference of Learning @ Scale (L@S '25). · Eason Chen, Xinyi Tang et al.
Starting Seatwork Earlier as a Valid Measure of Engagement(opens in new tab)
Student Learning
Proposes a novel engagement indicator (decision to begin seatwork earlier), validates it across systems and shows it reliably predicts math learning outcomes.
2025 · EDM 2025 · Ashish Gurung, Jionghao Lin et al.
Exploring the Benefit of Customizing Feedback Interventions for Educators and Students(opens in new tab)
Student Learning
Using contextual bandits to personalize tutor feedback delivery led to greater tutor engagement and improved student talk time metrics.
2025 · LAK 2025 · Joy Yun, Allen Nie et al.
Predicting Long-Term Student Outcomes from Short-Term EdTech Log Data(opens in new tab)
Student Learning
This study investigates the potential of using short-term interaction logs from educational technology to predict long-term student outcomes, such as end-of-year assessments. Analyzing data from diverse educational contexts, the findings suggest that 2-5 hours of initial usage can provide valuable insights into students' future performance.
2025 · 15th International Learning Analytics and Knowledge Conference (LAK2025) · Ge Gao, Amelia Leon et al.
Do Tutors Learn from Equity Training and Can Generative AI Assess It?(opens in new tab)
Gen AITutor Training
Evaluates equity training for tutors and explores using GPT-4 to assess tutors' responses to equity-related scenarios.
2024 · LAK 2025 · Danielle R. Thomas, Conrad Borchers et al.
Does Multiple Choice Have a Future in the Age of Generative AI?(opens in new tab)
Gen AITutor Training
MCQs are as effective and more efficient than open-response tasks for learning when practice time is limited.
2024 · LAK 2025 · Danielle R. Thomas, Conrad Borchers et al.
Examining the Use of an AI-Powered Teacher Orchestration Tool at Scale(opens in new tab)
Student LearningGen AI
How teachers perceive and use an AI-supported tool within a popular math platform to aid student learning.
2024 · L@S '24 · Emma Brunskill, Kole Norberg et al.
Learning and AI Evaluation of Tutors Responding to Students Engaging in Negative Self-Talk(opens in new tab)
Gen AITutor Training
GPT-4 can effectively evaluate tutor responses to student negative self-talk in online training.
2024 · L@S '24 · Danielle R. Thomas, Jionghao Lin et al.
How Can I Get It Right? Using GPT to Rephrase Incorrect Trainee Responses(opens in new tab)
Gen AITutor Training
GPT-4 provides timely, effective explanatory feedback in tutor training, achieving performance comparable to human experts.
2024 · International Journal of Artificial Intelligence in Education · Jionghao Lin, Zifei (Feifei) Han et al.
How Can I Improve? Using GPT to Highlight the Desired and Undesired Parts of Open-ended Responses(opens in new tab)
Gen AITutor Training
Fine-tuned GPT models can effectively identify components of praise in tutor training feedback.
2024 · EDM 2024 · Jionghao Lin, Eason Chen et al.
The Neglected 15%: Positive Effects of Hybrid Human-AI Tutoring Among Students with Disabilities(opens in new tab)
Student Learning
Hybrid human-AI tutoring significantly enhances learning and motivation for all students, particularly benefiting students with disabilities.
2024 · AIED 2024 · Danielle R. Thomas, Erin Gatz et al.
Using Generative AI to Provide Feedback to Adult Tutors in Training(opens in new tab)
Student LearningGen AITutor Training
Leveraging GPT-4 within the PLUS tutoring platform to provide real-time, explanatory feedback to adult tutors during scenario-based lessons.
2024 · The Learning Ideas Conference · Danielle R. Thomas, Erin Gatz et al.
Improving Student Learning with Hybrid Human-AI Tutoring: A Three-Study Investigation(opens in new tab)
Student LearningGen AI
Human-AI tutoring improves proficiency and engagement, especially for lower-achieving students in low-income schools.
2024 · LAK 2024 · Danielle R. Thomas, Jionghao Lin et al.
Improving Assessment of Tutoring Practices using Retrieval-Augmented Generation(opens in new tab)
Gen AITutor Training
Exploring GPT models with RAG prompting to assess novice tutors' social-emotional tutoring strategies.
2024 · AAAI 2024 Workshop on AI for Education · Zifei (Feifei) Han, Jionghao Lin et al.
Using Large Language Models to Assess Tutors' Performance in Reacting to Students Making Math Errors(opens in new tab)
Gen AITutor Training
GPT-3.5-Turbo and GPT-4 can effectively assess tutors' strategies for guiding students through math errors, though GPT-4 sometimes overidentifies errors.
2024 · AAAI2024 Workshop on AI for Education - Bridging Innovation and Responsibility · Sanjit Kakarla, Danielle R. Thomas et al.
Personalized Learning Squared (PLUS): Doubling Math Learning through AI-assisted Tutoring(opens in new tab)
Student LearningGen AITutor Training
PLUS combines human tutors and AI to enhance math learning for middle school students, especially those from diverse backgrounds.
2023 · AIED 2023 Workshop · Jionghao Lin, Danielle R. Thomas et al.