Predicting item difficulty using large language models
Combining large language models with tree-based machine learning to estimate assessment item difficulty.
International Journal of Artificial Intelligence in EducationML research & measurement scientist
I am a research scientist with expertise in machine learning, applied measurement, and behavioral science. My research interests center on AI evaluation and human-centered AI.
Large language models and human-centered AI.
Psychometrics and applied measurement.
Predictive modeling and statistical learning.
Experimental design and empirical research methodology.
Combining large language models with tree-based machine learning to estimate assessment item difficulty.
International Journal of Artificial Intelligence in EducationExamining the ability of large language models to identify interpersonal emotion regulation strategies from narratives.
EmotionCombining structural topic modeling and dictionary-based text analysis with multidimensional measurement to study justified and unjustified anger.
Collabra: PsychologyFirst-place results in two consecutive rounds of the Center for Open Science’s Predicting Replicability Challenge.
Awards & recognitionFirst place
AI/ML Competition · Round 2
Center for Open Science
First place
AI/ML Competition · Round 1
Center for Open Science
Evaluating an AI system involves more than assigning a score. The interpretation of any evaluation depends on what is being measured, how it is measured, and what conclusions the resulting evidence supports. Decades of research in psychometrics and behavioral science offer an important foundation for the evaluation and development of AI systems.
Background & expertiseI welcome scientific conversations and professional inquiries.