Skip to content / 跳至主要內容
Technology & research

When should a teacher step in?

CMU's Lumilo research offers a way to think about persistent difficulty and the evidence teachers need before intervening.

One wrong answer is not the whole signal

A student may be making a useful attempt or repeatedly getting stuck. Intervention signals should include attempts and help-seeking behavior so teachers can judge whether practice is progressing or a different explanation is needed.

What CMU's Lumilo research contributes

Holstein, McLaren and Aleven's AIED 2018 study examines Lumilo, a teacher-awareness tool in AI-enhanced classrooms. It presents signals such as unproductive persistence and patterns of help use to support teacher–AI coordination.

The principle relevant to Vispo is to pair an alert with interpretable evidence and preserve teacher judgment. This is a discussion of related research, not a CMU endorsement, partnership or validated Vispo outcome.

Make an alert useful to the teacher

A useful handoff explains the task, the student's attempts, what remains unclear after a hint and a possible next step. Observed evidence and AI inference should appear separately.

Alert volume, timely teacher access and learning after intervention need evaluation in real teaching settings.

References

  1. Holstein, McLaren & Aleven · Student Learning Benefits of a Mixed-Reality Teacher Awareness Tool in AI-Enhanced Classrooms (AIED 2018)
Read more technical notes
Vispo|智加雲 AI 教學助手