Instructing Llama 3.1 with monitoring and scaffolding rules increases question-asking and perceived co-constructiveness in explanation dialogues, but not average objective understanding.
Figure 7: The questionnaire to assess the participants’ prior subjective comprehension of sleep cylces and the motivation to learn about it before interacting with the LLM
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Investigating Co-Constructive Behavior of Large Language Models in Explanation Dialogues
Instructing Llama 3.1 with monitoring and scaffolding rules increases question-asking and perceived co-constructiveness in explanation dialogues, but not average objective understanding.