DALL-E 3 disproportionately generates animals matching cultural stereotypes when prompted with trait adjectives, and a simple anti-stereotyping instruction only partially mitigates this.
Firstly, the analysis was restricted to a small set of prompts representing specific traits, which may not capture the full range of biases present in the model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Owls are wise and foxes are unfaithful: Uncovering animal stereotypes in vision-language models
DALL-E 3 disproportionately generates animals matching cultural stereotypes when prompted with trait adjectives, and a simple anti-stereotyping instruction only partially mitigates this.