When LLM Tutoring Responses Work: Evidence from Student Programming Conversations
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/2607.09919.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryAnalyzes 16,851 LLM tutoring responses from 203 students in a CS course, finding response style significantly associates with productive continuation: verification feedback yielded the highest rate (82.4%), direct answers the lowest (62.7%), though effect sizes were small and context-dependent.
Why it mattersConcrete evidence that verification-style feedback outperforms direct answers in LLM programming tutoring, with guidance on matching response style to student cognitive load.
Cited by
No citations on record.