Showing cs.CLShow all
3 papers · 1 filter
cs.CL2026
Detecting Knowledge Gaps from Conversational AI Interactions Using Curriculum Prerequisite Graphs
Youssef Medhat, Junsoo Park, Ploy Thajchayapong +1
Large online courses generate thousands of student questions directed at conversational AI teaching assistants, yet these interaction logs remain largely untapped as diagnostic sig…
cs.CL2026
Evaluating Learner Representations for Differentiation Prior to Instructional Outcomes
Junsoo Park, Youssef Medhat, Htet Phyo Wai +2
Learner representations play a central role in educational AI systems, yet it is often unclear whether they preserve meaningful differences between students when instructional outc…
cs.CL2024
OffsetBias: Leveraging Debiased Data for Tuning Evaluators
Junsoo Park, Seungyeon Jwa, Meiying Ren +2
Employing Large Language Models (LLMs) to assess the quality of generated responses, such as prompting instruct-tuned models or fine-tuning judge models, has become a widely adopte…