LLMs show a causal illusion bias: they often report causal relationships when evidence is purely correlational, null-contingency, or temporally impossible, particularly in 0-100 scaled judgments.
Hamilton and Robert K
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Do Large Language Models Show Biases in Causal Learning?
LLMs show a causal illusion bias: they often report causal relationships when evidence is purely correlational, null-contingency, or temporally impossible, particularly in 0-100 scaled judgments.