EduPanel: A Three-Agent LLM Judge for Teaching Videos -- Reliability, Complementarity, and Human Trust Calibration
3.00T1 sourcearXiv cs.HC
arXiv cs.HCoriginal source ↗
Source record
Published by arXiv cs.HC (T1 source). The original is at https://arxiv.org/abs/2607.18529.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryEduPanel is a rubric-grounded, learner-conditioned LLM judge using three specialized agents to evaluate teaching video quality. Expert studies report reliability matching a median human expert, improved scoring accuracy, and expert ability to flag unreliable outputs via AUC 0.77.
Why it mattersConcrete three-agent decomposition pattern with measured human trust calibration, not a blanket 'replace humans' claim.
Cited by
No citations on record.