Simulated students that make realistic mistakes help AI tutors learn faster

Microsoft and the University of Illinois built StudentSim to replicate individual students from limited data and give AI tutors fast, low-cost feedback. In tests covering 60 students across chess, English, and math, it outperformed GPT-5.4. A chess tutor trained with StudentSim also earned the…

Microsoft and the University of Illinois built StudentSim to replicate individual students from limited data and give AI tutors fast, low-cost feedback. In tests covering 60 students across chess, English, and math, it outperformed GPT-5.4. A chess tutor trained with StudentSim also earned the highest expert ratings among three versions tested. The article Simulated students that make realistic mistakes help AI tutors learn faster appeared first on The Decoder.

Source: The Decoder — Published September 20, 2026 at 09:50 UTC — Category: Models