AI RESEARCH

Catching The Correct Answer Trap: Characterising AI Tutor Blind Spots When Analysing Student Reasoning

arXiv CS.AI

ArXi:2605.23925v1 Announce Type: cross Intelligent tutoring systems increasingly provide automated feedback on student work, but robust feedback requires assessing reasoning, not only final answers. We study a failure mode we call the correct answer trap (CAT): models under-detect misconceptions when students reach a correct answer via flawed reasoning.