Microsoft Research releases StudentSim, a student simulator for testing AI tutors

Microsoft Research introduces StudentSim, a research system that builds personalized AI simulators of individual students to test AI tutoring tools before they are used with real learners.

According to the research announcement, the simulators model how each student answers questions and how those answers change after tutor guidance. The paper and the StudentSim training and evaluation code, including the StudentSimEval benchmark, are publicly available. StudentSimEval covers chess, second-language English writing, and mathematics, and the simulators are scored on how closely they match a specific student's answers and how they respond to tutor feedback. The announcement reports no test results and names no tutoring product that has used the system.