Multimodal LLMs systematically answer forced-choice social questions differently for beautified versus original faces, assigning positive traits to the beautified versions in most scenarios.
Cash and Robert N
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
-
Beauty and the Bias: Exploring the Impact of Attractiveness on Multimodal Large Language Models
Multimodal LLMs systematically answer forced-choice social questions differently for beautified versus original faces, assigning positive traits to the beautified versions in most scenarios.