A 14-week deployment of a course-specific AI teaching assistant shows frequent users rated it more helpful and comfortable over time, though trust stayed below human instructors.
How much would you trust the responses from an LLM-based TA?
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Large-Scale Real-World Evaluation of LLM-Based Virtual Teaching Assistant
A 14-week deployment of a course-specific AI teaching assistant shows frequent users rated it more helpful and comfortable over time, though trust stayed below human instructors.