Generate targeted diagnostic follow-up questions using variant-problem and probe-the-gap strategies for ambiguous cases¶

Strategy · All strategies
Evidence · no claims cited

Description¶

For correct-answer cases the model routes to needs-clarification, the system generates a targeted diagnostic follow-up question using either a variant-problem strategy, presenting a near-transfer problem where the suspected misconception would produce the wrong answer, or a probe-the-gap strategy, asking about the specific step the student omitted. The paper notes that in autonomous tutor mode, "asking a student to explain their reasoning is good formative practice regardless of whether a misconception is present", making false positives pedagogically benign while false negatives leave misconceptions unchallenged.

Design Implications¶

Context¶

Requirements¶

• A reasoning-capable LLM generating the follow-up question and a routed ambiguous case from the graduated assessment

Constraints¶

• The proof of concept generated follow-up questions for 7 TM instances from the test set; no student responses to these questions have been collected, and a classroom trial would be needed to validate the pipeline

Target Learners¶

• school mathematics students whose explanations are too brief to diagnose

Target Learning Goals¶

• eliciting evidence of student reasoning through formative follow-up questioning

Related Strategies¶

• Probing Questions
• Describe a support system's design intent by asking against whose Synthesis or Analysis it elicits a structural gap

Examples¶

-

Key Sources¶

• Moiz Imran, Sahan Bulathwela. (2026). The Correct Answer Trap: Pedagogically-Grounded Detection and Feedback for Hidden Misconceptions. PEAF 2026: Workshop on Pedagogical Evaluation of Automated Feedback. https://orcid.org/0000-0002-5878-2143