LLMs show stronger knowledge-seeking curiosity than humans but make more conservative choices in uncertain environments, and curiosity-like behaviors improve their reasoning and active learning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Why Did Apple Fall: Evaluating Curiosity in Large Language Models
LLMs show stronger knowledge-seeking curiosity than humans but make more conservative choices in uncertain environments, and curiosity-like behaviors improve their reasoning and active learning.